Effect of Non-ionic Surfactants and Its Role in K Intercalation

in Electrolytic Manganese Dioxide

AVIJIT BISWAL, B.C. TRIPATHY, T. SUBBAIAH, D. MEYRICK, MIHAIL IONESCU,
and MANICKAM MINAKSHI

The effect of non-ionic surface active agents (surfactants) Triton X-100 (TX-100) and Tween-20
(Tw-20) and their role in potassium intercalation in electrolytic manganese dioxide (EMD)
produced from manganese cake has been investigated. Electrosynthesis of MnO2 in the absence
or presence of surfactant was carried out from acidic MnSO4 solution obtained from manganese
cake under optimized conditions. A range of characterization techniques, including field
emission scanning electron microscopy, transmission electron microscopy (TEM), Rutherford
back scattering (RBS), and BET surface area/porosity studies, was carried out to determine the
structural and chemical characteristics of the EMD. Galvanostatic (discharge) and potentio-
static (cyclic voltammetric) studies were employed to evaluate the suitability of EMD in com-
bination with KOH electrolyte for alkaline battery applications. The presence of surfactant
played an important role in modifying the physicochemical properties of the EMD by increasing
the surface area of the material and hence, enhancing its electrochemical performance. The
TEM and RBS analyses of the discharged EMD (c-MnO2) material showed clear evidence of
potassium intercalation or at least the formation of a film on the MnO2 surface. The extent of
intercalation was greater for EMD deposited in the presence of TX-100. Discharged MnO2
showed products of Mn2+ intermediates such as MnOOH and Mn3O4.

DOI: 10.1007/s40553-014-0022-9
 ASM International (ASM) and The Minerals, Metals & Materials Society (TMS) 2014

I. INTRODUCTION material derived from EMD. This rechargeable battery

THE intense interest in manganese oxides for energy
applications is driven by the low cost and non-toxicity of
these materials when compared with nickel and cobalt
oxides. The rich chemistry and diverse crystalline
structures of MnO2 offer a versatility that allows its
use in a range of applications.[1] Of the different types of
MnO2, such as natural manganese dioxide (NMD),
chemical manganese dioxide (CMD), activated manga-
nese dioxide (AMD), and electrolytic manganese diox-
ide (EMD), EMD is most widely applied in battery
industries.[2,3] EMD is composed predominantly of the
c-phase and may be produced cost effectively with low
environmental impact. EMD has a high redox potential,
high rate capability, long storage life, and better
performance over a wide temperature range relative to
its nickel and cobalt counterparts.[4–8]
It is anticipated that over the next decade, the pricing for
high purity EMD will increasingly be affected by the
demand for lithium-ion batteries containing cathodic
AVIJIT BISWAL, Ph.D. Scholar, and B.C. TRIPATHY and T.
SUBBAIAH, Senior Scientists, are with the CSIR - Institute of Minerals
and Materials Technology, Bhubaneswar 751013, India, and also with
the Academy of Scientific and Innovative Research, Anusandhan
Bhavan, New Delhi 110001. D. MEYRICK, Senior Lecturer, and
MANICKAM MINAKSHI, Lecturer, are with the School of Engineer-
ing and Information Technology, Murdoch University, Murdoch, WA
6150, Australia. Contact e-mail: minakshi@murdoch.edu.au MIHAIL
IONESCU, Senior Research Scientist, is with the Environment, Lucas
Heights Laboratories, ANSTO, Menai, NSW 2234, Australia.
Manuscript submitted March 21, 2014.
has quickly become the technology of choice in the cordless
power tool industry and is being increasingly applied in the
electric vehicle (EV) and hybrid electric vehicle (HEV)
industries. The EV and HEV battery markets will likely
expand rapidly over the next few years, and this will
translate to attractive returns for high-end EMD producers.
An estimated two billion manganese versions of lithium
primary (non-rechargeable) batteries and half a billion
lithium manganese secondary (rechargeable) batteries are
produced per annum, with growth rates in excess of 10 and
20 pct, respectively.[9] This increasing demand gives rise to a
need for high purity EMD from alternative sources.
Numerous reports are available regarding the synthesis of
EMD (c-MnO2) from various manganese ores[10–14] and its
role as a cathode in alkaline Zn-MnO2 systems,[3,4,8,15] while
the possibility of producing EMD from low grade and
secondary manganese resources has not been fully ex-
plored. Production of EMD from a range of sources will be
necessary to meet future escalating demand.
In this context, we have produced battery grade
EMD, in the presence of surfactants, from manganese
cake, a side product obtained during the processing of
manganese nodules.[16] Surfactants play a significant
role in modifying the growth pattern (and hence the
physicochemical properties) of electrodeposits through
adsorption on the MnO2 electrode surface.[17] Surface
adsorption of surfactants influences the kinetics of
electron transfer through blocking of active sites and
also affects electrostatic interactions between electroac-
tive species in the electrolytic bath. Consequently,
addition of organic surfactants to the electrolytic bath

METALLURGICAL AND MATERIALS TRANSACTIONS E

affects the morphology and mechanical properties of
electrochemically deposited material, leading to altered
electrochemical behavior of materials. Use of surfac-
tants during preparation of materials for battery appli-
cations has been widely reported in the literature.[17–23]
Surfactants may be used during chemical co-precipita-
tion,[18,19] liquid co-precipitation,[20] as well as electro-
chemical deposition methods.[21,22] Ghaemi et al.[22]
have reported improved charge/discharge cycle behavior
of EMD prepared in the presence of t-octyl phenoxy
polyethoxy ethanol (Triton X-100). The absence of this
surfactant was found to produce EMD unsuitable for a
rechargeable battery application. EMD prepared in the
presence of cetyltrimethylammonium bromide (CTAB)
or sodium n-dodecylbenzenesulfonate (SDBS) has also
been reported and found suitable for battery applica-
tions.[23] The positive effect of use of CTAB and SDBS
on the performance of Zn-MnO2 alkaline batteries has
been reported earlier.[23] Triton X-100 is reported[21] to
offer the best improvement in capacitance behavior of
EMD, possibly due to strong adsorption of the surfac-
tant, resulting in an increase in specific capacitance of
the energy storage device.
The current work evaluates the use of Triton X-100
along with Tween-20 on EMD produced from a
secondary source. The difference in chemical structure
observed (detailed in results and discussion section) for
the two non-ionic surfactants (TX100 and Tw20)
intrigued to investigate their role in EMD for energy
storage devices. To the best of our knowledge, no work
has been reported on the suitability for rechargeable
battery applications of EMD from manganese cake
prepared in the presence of the above mentioned
surfactants. Most of the reported literature on EMD
material in aqueous solutions is based on the well-
known protonation or lithiation mechanisms.[24–28] In
this work, TEM imaging and Rutherford back scatter-
ing analyses gave intriguing insights suggesting that the
potassium ion may be intercalated in the c-MnO2
structure. The extent of intercalation depends on the
type of surfactant used. The influence of these two
surfactants on the structural, morphological, and elec-
trochemical characteristics of the EMD produced from
purified manganese sulfate solution obtained from
manganese cake was thus investigated. This work not
only examines the suitability of these two surfactants in
improving the electrochemical activity of MnO2 material
but also opens a new possibility for utilization of a
secondary source to produce an energy material.
II. EXPERIMENTAL
A. Materials and Methods for Producing EMD
Electrolytic manganese dioxide (EMD) was produced
from purified MnSO4 solutions obtained from manga-
nese cake at an anodic current density of 200 Am2 in a
glass cell, as reported in our previous work.[16] Purified
MnSO4 solution was obtained after leaching of manga-
nese cake, followed by two stages of purification.[16,29]
All EMD samples were prepared with in situ addition of
a calculated (as per critical micelle concentration)
quantity of surfactant to the electrolytic bath containing
50 g dm3 MnSO4 and 25 g dm3 H2SO4. The critical
micelle concentrations (CMC) of Triton X-100 (TX-100,
SLR) and Tween-20 (Tw-20, SLR) are 0.02 and
0.01 wt pct, respectively. EMD samples prepared from
0 t, 5 t, 10 t, 15 t, and 30 t cmc values of TX-100 (where
t denotes CMC) have been labeled EMD0, EMDTX5,
EMDTX10, EMDTX15, and EMDTX30, respectively. Sim-
ilarly the EMD produced from the electrolytic bath
containing 10, 15 t, and 30 t CMC of Tw-20 has been
labeled EMDTw10, EMDTw15, and EMDTw30, respec-
tively. The electro-synthesized EMD was scraped off
from the anodic substrate after the deposition and
washed thoroughly with deionized water before drying
in an oven. The dried mass was ground and sieved
through a 53-lm mesh to obtain EMD powder. The
resultant product in powder form was washed repeat-
edly with de-ionized water, until the sample was sulfate
free. The EMD powder was finally dried and cooled in a
desiccator and subjected to physical and electrochemical
characterization.
B. Structural, Thermal, and Microscopic Analyses
X-ray diffractograms were recorded for the resultant
EMD powders using PANalytical diffractometer (PW
1830, Philips, Japan) with Cu Ka radiation, k = 1.5418 Å.
The scans were recorded in 2h range 15 to 80 deg. BET
surface area and pore size analysis were carried out using
Quantachrome (Autosorb-iQ) Surface area analyzer. Dif-
ferential thermal analysis (DTA) and thermogravimetry
(TG) (Perkin Elmer Diamond) of the EMD samples were
carried out under an inert atmosphere over a temperature
range between 303 K and 1273 K (30 C and 1000 C)
with a heating rate of 278 K (5 C) min1. The surface
morphology of the EMD samples was determined using
field emission scanning electron microscope (FESEM,
ZEISS: SUPRA55). Fourier transform infrared (FT-IR)
spectra were recorded on a Nicolet 6070 spectrophotom-
eter in the frequency range 400 to 4500 cm1. The
morphology and interplanar spacings of the products
formed before and after discharge were determined by
transmission electron microscopy (TEM) coupled with
energy dispersive analysis (EDS) using a JEOL 2010F
TEM model operated at 200 kV. TEM specimens were
prepared by grinding a small fragment of material scraped
from the pressed pellet under methanol and dispersing on a
perforated carbon support film. Specimens were examined
at liquid nitrogen temperature in a cooling stage, to reduce
beam damage and contamination effects. Rutherford back
scattering analysis (RBS) was used in conjunction with
proton induced X-ray emission (PIXE) to determine the
concentrations of elements such as O, C, Na, Ni, Co, Mn,
and P. The RBS and PIXE measurements were made
simultaneously on an ion beam accelerator using a
2.6 MeV proton beam at ANSTO facilities.
METALLURGICAL AND MATERIALS TRANSACTIONS E
C. Electrochemical Characterization
The pellet preparation, cell arrangement, and all other
conditions and parameters were identical to those of our
previous work.[29] For cyclic voltammetric (CV) experi-
ments, a standard three-electrode cell was used. For this
purpose, EMD was used as the working electrode. The
counter electrode was a zinc foil, which was separated from
the main electrolyte (9 M KOH) by means of a porous frit
to avoid the spallation of zinc inserted into the MnO2.
Mercury-mercury oxide (Hg/HgO) served as the reference
electrode. Reported potentials are relative to Hg/HgO
with a slow scan rate of 50 lV/s. For the galvanostatic
experiments the applied discharge current was 20 mA
with a cut-off voltage (COV) of 0.9 V, while the applied
charge current was 40 mA with a cut-off voltage of 1.9 V.
The discharge capacities were recorded for up to 15 cycles.
The galvanostatic measurements were carried out using a
BITRODE deep cycle battery tester (LCN1-25-24, USA).
III. RESULTS AND DISCUSSION
A. Electrolysis
The current efficiency (CE) and energy consumption
(EC) of electrodeposited EMD prepared in the presence of
TX-100 and Tw-20 surfactants are given in Table I. In the
absence of surfactant as additive in the bath, 89 pct CE was
observed. With the introduction of a small amount of
surfactant (TX-100), the CE slightly increased to 89.5 pct
with slight decrease in EC from 1.65 to 1.64 kWh kg1.
With an increase in the concentration of TX-100, the
current efficiency decreased, and an increase in energy
consumption was observed, while with the addition of ten
times the CMC of Tw-20, the CE increased to 98 pct, while
EC decreased from 1.65 to 1.46 kWh kg1 during elec-
trolysis. This suggests strong adsorption of the additives on
the electrode surface, as suggested by Ghaemi et al.[22]
In general, the electrodeposition of manganese diox-
ide from an acidic sulfate solution proceeds through the
following reactions:
At anode: Mn2þ þ 2H2 O ! MnO2 þ 4Hþ þ 2e ½1
At cathode: 2Hþ þ 2e ! H2 : ½2
At the anode, the formation of MnO2 does not take
place in a single step; rather, Mn3+ as an intermediate
Table I. Effect of the Surfactants TX-100 and Tw-20 on Current Efficiency (CE) and Energy Consumption (EC) during Electrode-
position of MnO2
Amount
Surfactant in CMC CE (pct) EC (kWh kg1)
Nil 89 1.65
TX 10 t 89.5 1.64
15 t 87 1.74
30 t 70 2.07
Tw 10 t 98.3 1.46
15 t 95.7 1.58
30 t 85.8 1.72
METALLURGICAL AND MATERIALS TRANSACTIONS E
species is first formed[24,26,30–32] together with some solid
intermediates such as MnOOH(s) and Mn2O3(s). Being
unstable in acidic solution,[2] the Mn3+ ion undergoes a
disproportionation reaction forming Mn4+ and Mn2+.
Mn2þ 3þ 
sol ! Mnads þ e ½3
2Mn3þ 4þ 2þ
ads ! Mnads þ Mnsol : ½4
During the electrodeposition process, Mn3+ ions may
be trapped in the MnO2 lattice, possibly resulting in
defects in the crystal structure.[22] Adsorption of surfac-
tants at the substrate/electrolytic solution interface may
inhibit the rate of Eq. [3] due to the blocking of the
active growth sites, thereby allowing electrodeposition
preferentially on the crevices. The electron/ion transfer
kinetics depends largely on the degree of surface
coverage of the electrode due to mechanical blocking
that is caused through electrostatic interactions.[33,34]
Coverage of the electrode surface may change the
electrical double layer characteristics and thus affect
the interfacial energy, dielectric constant, potential, and
current distribution at the electrodes, resulting in
modified crystal growth. Hence, organic surfactants
may play a crucial role in facilitating the formation of
compact deposits with greater surface area. However,
the effect of surfactant concentration should be well
understood. Higher concentrations may lead to irregular
morphology owing to higher ohmic potential drop and
electrode overvoltage.[35]
B. Structure and Function of Surfactant in the Solution
Surfactants lower the surface tension of a liquid as
well as liquid–liquid interfacial tension or liquid–solid
interfacial tension. Surfactants are amphiphilic mole-
cules with both hydrophilic and lipophilic moieties. The
adsorption and desorption of these amphiphilic mole-
cules at the solid liquid interface were possible in this
work with the given applied potential range. Critical
micelle concentration (CMC) is the concentration of a
surfactant above which micelle formation occurs. The
existence of micelles does not preclude the existence of
individual surfactant molecules in solution. Further
addition of surfactant after reaching the CMC will
ultimately increase the no. of micelles. The structure of a
material electrodeposited in the presence of a surfactant
BET Surface Area (m2 g1) Open Circuit Voltage (V)
101.908 1.550
125.843 1.605
173.982 1.578
114.685 1.482
146.971 1.548
152.525 1.508
133.188 1.480
can be related to the micelle population. With an
increase in the concentration of surfactant (from 10 to e
30 t in the current study), a different morphological
pattern may be created in the diffusion layer of the
electrode.[33] Surfactant adsorbed to charged surfaces d
may arrange in bilayers called admicelles, forming a
molecular layer with head groups down on the surface

Intensity (a.u)
and a second layer with head groups facing the c
solution.[36] The CMC values reported for non-ionic
surfactants Tw-20 and TX-100 are 8.5 9 104 and
9.2 9 104 M, respectively.[37] The surfactant with low- b
er CMC implies higher adsorbed amounts. Hence, to 110
make a comparative study of the effect of surfactants on 021 240 002
121 221
the electrochemical processes, the two non-ionic surfac- 061 a
tants (Tw and TX) with different chemical structures
were chosen.
In the case of TX-100, the hydrophobic aromatic 15 20 25 30 35 40 45 50 55 60 65 70 75 80
hydrocarbon is oriented toward the solution.[34] With an 2θ (degrees)
increase in surfactant concentration (beyond the CMC),
surfactants may assume a range of conformations Fig. 1—X-ray diffraction patterns of (a) EMD0, (b) EMDTX15, (c)
including the interweaving of hydrophobic chains of EMDTX30, (d) EMDTw15, (e) EMDTw30.
adjacent molecules and the formation of admicelles.[38]
The underlying mechanism of adsorption is an attrac- 50
tion between the oxygen atoms in polyoxyethylene MnO2 region
45
groups of TX-100 or Tw molecule and the Mn+2
species, forming Mn+2-TX and Mn+2-Tw pairs, respec- 40
Transmittance / a.u.

EMDTX30
tively. These pairs will be affected by surfactant con- 35
centration and will influence the electrodeposition of
30 EMDTW30
MnO2. Hence, the concentration of surfactant should be
limited to a particular value, above which there will be 25 EMDTW15
undesirable consequences in terms of physical and 20
electrochemical characteristics of deposited EMD. EMDTX15
15
1420
10 EMD0
C. Structural Characterization 1635 1130
5 3375
Figure 1 shows the X-ray diffraction (XRD) patterns 754
of EMD samples prepared at different concentrations of 0
4000 3500 3000 2500 2000 1500 1000 500
surfactants. The diffraction patterns show in situ addi-
-1
tion of either surfactant to the electrolytic bath had no Wavenumber / Cm
significant effect on the crystalline phase of c-MnO2. The
peak intensity for all samples is weak, with peak Fig. 2—FTIR spectra of EMD samples (a) EMD0, (b) EMDTX15, (c)
broadening indicating a typical amorphous quality to EMDTw15, (d) EMDTw30, (e) EMDTX30.
the deposit. All of the diffraction peaks can be indexed
to an orthorhombic phase of c-MnO2 with lattice with the water crystallization,[40] and a broad absorption
constants: a = 8.70 Å, b = 2.90 Å and c = 4.41 Å. band at ~3400 cm1 is due to OH stretching vibra-
This is in good agreement with the standard values tions.[40–43] Additional peaks at 968 and 1025 cm1 are
(JCPDS card no. 65-1298; a = 9.27 Å, b = 2.87 Å, attributed to Mn-O-H structural vibrations in agree-
c = 4.53 Å). The peaks at 2h values of ~22, ~37.2, ~42, ment with the spectral features reported by McBride.[44]
~56, and ~68 deg correspond to the 110, 021, 121, 240/
221, and 002/061 crystal planes of c-MnO2, respectively. D. Thermal Analysis
The FTIR spectra recorded for the EMD samples in
the presence and absence of surfactant (Figure 2) show Thermodynamic stability as well as electrical conduc-
significant IR peaks at different frequency ranges. The tivity of manganese dioxide are strongly governed by the
broad absorption band detected in the finger print presence of water molecules in the crystal lattices; the
region of 400 to 600 cm1 confirms the structural water content causes a variation in the crystal lattice
vibrations occur in MnO6 octahedral in c-MnO2 struc- with a consequent effect on electrical conductivity. The
ture.[39] The absorption band at 1100 cm1 can be water content enhances the transportation of active
attributed to the MnO2 stretching mode and/or O-H ionic species and consequently increases the specific
bending vibrations[39] associated with hydrogen bond- capacitance of manganese dioxide.[45] Hence to deter-
ing, indicating the presence of bound water mole- mine the role of thermal stability, thermal analysis
cules.[40,41] The strong absorption band at ~1630 cm1 (TGA and DTA) was carried out for all EMD samples
can be attributed to O-H bending vibrations associated (Figures 3(a) and (b)). Release of physisorbed water

METALLURGICAL AND MATERIALS TRANSACTIONS E

below 383 K (110 C) is observed for all samples. The
mass loss in the region of 383 K to 573 K (110 C to
300 C) is due to removal of chemically bound water
associated with c-MnO2. This is followed by mass loss in
the temperature range 673 K to 873 K (400 C to
600 C) and 873 K to 1173 K (600 C to 900 C), which
indicates thermal decomposition resulting in the forma-
tion of Mn2O3 and Mn3O4, respectively.[46,47] Among
the doped and undoped sample, the mass loss during
this step is substantially higher for EMDTX30, EMD0,
and EMDTw30. However, EMDTX15 and EMDTw15
showed quite stable behavior on thermal analysis
throughout the temperature range.
Differential Thermal Analysis (DTA) shown in
Figure 3(b) indicates that all EMD samples in this work
show a sharp intense peak around 100 C corresponding
to the loss of physisorbed water. A broad endo peak at
~543 K (~270 C) for EMDTX30 and at 578 K (305 C)
for EMD0 corresponds to the removal of all the
structural water. This peak was absent in other EMD
samples. The sharp endo peak observed at (833 K)
560 C for EMD sample corresponds to the thermal
34
32
30
Weight loss / %
decomposition of MnO2 to Mn2O3,[48] and is insignif-
icant in case of EMDTX15 (at ~768 K (~495 C) and
EMDTw15 (at ~770 K (~497 C)). Endo peaks observed
~1173 K (~900 C) attributed to the decomposition of
Mn2O3 to Mn3O4. From the TG and DTA data, it may
be concluded that the more significant water loss in the
early stage of heating for EMDTX30, EMDTw30, and
EMD may be due to a loose structure with a greater
quantity of interlayer water. This corresponds to the
instability in the structure responsible for the deteriora-
tion of these EMD samples during reduction–oxidation
in battery applications. However, EMDTX15 and
EMDTw15 showed less intense peaks, suggesting that
bound water molecules responsible for the electrochem-
ical activity of EMD are more strongly bound in these
EMD samples, imparting structural stability to the
material.
E. Surface Morphology
Surfactants may play an important role in modifying
the structure of EMD by controlling the nucleation and
growth mechanism during electrodeposition[49] pro-
cesses, which in turn impinge on the stability and
rechargeability of EMD.[22] The presence of surfactant
in the solution strongly affects the particle size of EMD
material. In this work, smaller grains have been formed

28
in both EMDTX15 and EMDTw15 samples (Figures 4(a)
and (b)) relative to those obtained from EMD produced
26 in the surfactant-free bath. However, with increase in
(e)
24 surfactant concentration (in EMDTX30 and EMDTw30),
(d)
22 the grain size increases. All the EMD samples consist of
(c) particles with an angular shape having a rough surface.
20
On magnifying the surface of the EMD grain, it was
18 (b) observed that EMD0 seems to consist of nanopores,
16 (a) while EMDTX15 and EMDTX30 show fibrous- and spike-
like growth, respectively. EMDTw15 consists of spherical
0 200 400 600 800 1000
nanoparticles in the range of 300 nm with an agglom-
Temperature / °C erated form.
(a)
F. Surface Area and Pore Size
(e)
4 Surface area and porosity play significant roles in
controlling the electrochemical activity of EMD. Sur-
(d)
(c)
face area reached a maximum at 15 t for each of the
2
surfactants (Table I) for the EMD produced in this
Weight loss / %

(b)
work. Pore size distribution of the EMD samples was
0 carried out to evaluate its effect on rate capability of
(a) EMD produced in the presence of surfactant. All EMD
-2 samples showed isotherms of type-IV pattern with
strong hysteresis behavior. The adsorption–desorption
-4 isotherm as well as the shape and position of the
hysteresis loop were affected by the morphology and
specific pore size.
-6
0 200 400 600 800 1000 Figure 5 shows the nitrogen adsorption–desorption
Temperature / °C isotherm of the doped (presence of surfactants) and
undoped (absence of surfactants) EMD samples. The
(b) hysteresis loops obtained for both EMD0 and EMDTX30
are a hybrid of type H2, H3, and H4; however,
Fig. 3—(a) TGA curves of (a) EMDTX15, (b) EMDTw15, (c)
EMDTw30, (d) EMD0, and (e) EMDTX30. (b) DTA curves of (a)
EMDTX15, EMDTw15, and EMDTw30 show the behavior
EMD0, (b) EMDTX15, (c) EMDTX30, (d) EMDTw15, and (e) of H2 type hysteresis loop with a higher surface area
EMDTw30. relative to the EMD0 and EMDTX30. The wide hysteresis

METALLURGICAL AND MATERIALS TRANSACTIONS E

Fig. 4—a FESEM images of EMD0 (a1, a2), EMDTX15 (b1, b2), EMDTX30 (c1, c2). b FESEM images of EMDTw15 (d1, d2), EMDTw30 (e1, e2).
loop indicates that all the EMD samples are predom-
inantly composed of mesopores. Mesopores with radius
of 1.87 to 1.97 and 1.68 to 1.95 nm are obtained for
EMDTX15 and EMDTw15, respectively, versus EMD0
with a pore radius of 1.58 to 1.96 nm. The pore size
diameters and distribution of all samples were calculated
using the Barrett–Joyner–Halenda (BJH) method. It
was also observed that with addition of surfactant, the
surface area of EMD increases markedly (Table I),
limited by a certain concentration (CMC value) of
surfactant, after which it decreases. EMDTX15 and
EMDTw15 show the highest surface area of 173.9 and
152.5 m2 g1 in comparison with EMD0, which shows a
surface area of 101.9 m2 g1. In general, EMD prepared
in the presence of surfactant shows a higher surface area
than that produced in the absence of surfactant. As
METALLURGICAL AND MATERIALS TRANSACTIONS E
Fig. 4—continued.
reported by[50,51] charge transfer processes between the
electrolyte and the crystal phase can be affected by
micro/meso porosity, and a decrease in the pore size
may lead to reduction of accessible electrochemical
active sites. To support this fact, doped EMD samples
obtained in this work with high surface area as well as
with greater pore size show improved cyclic behavior
(discussed in the Section III–G) relative to ‘‘non-surfac-
tant’’ material, due to greater electrochemically accessi-
bility of sites available for proton diffusion inside MnO2
cathode.
G. Electrochemical Activity
The electrochemical behavior of the EMD samples was
characterized by potentiostatic (potential controlled) and
galvanostatic (current controlled) experiments. Cyclic
voltammograms of the doped and undoped EMD mate-
rials are shown in Figure 6. The cyclic voltammogram of
EMD0 (Figure 6(a)) is initiated in the cathodic region and
then reversed back to the anodic region before it finishes
METALLURGICAL AND MATERIALS TRANSACTIONS E
at the starting potential. The CV is characterized by a
well-defined reduction peak (C1 = 0.390 V) and an
oxidation peak (A1 = 0.09 V) which is a fingerprint for
electron transfer processes corresponding to a redox
behavior suitable for energy storage and battery applica-
tions. For TX-added EMD, it is observed that the
cathodic and anodic currents in Figure 6(b) are higher,
and the reduction and oxidation peaks (at C1 = 0.450
and A1 = 0.030) have become more defined. The effect of
adding more TX (TX30) during electrodeposition of
EMD also had an impact on the voltammogram shown in
Figure 6(c). The redox peaks have shifted to lower
potentials (at C1 = 0.430 and A1 = 0.025) with the
separation between the peaks reduced to 0.040 V, indi-
cating excellent reversibility. However, the area under the
peak for TX30 is less than that for TX15. Further
addition of TX did not influence the redox process (not
shown here). TX has both hydrophilic and lipophilic
groups so that when its concentration reaches or sur-
passes CMC, the surfactant tends to form micelles, and
the adsorption of Mn2+ ions would be comparatively
 70

60

Volume (ccg STP)

50

-1
40

30 adsorption
desorption
20

0.0 0.2 0.4 0.6 0.8 1.0

Relative pressure ( P/P 0 )
EMD0

60 70
55
60
50
Volume ( ccg-1 STP )

Volume( ccg-1 STP)

45 50
40
40
35
30 30
25 adsorption adsorption
desorption 20
20 desorption
15 10
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Relative pressure ( P/ P0) Relative pressure (P/P0)
EMD TX15 EMD TX30
45
55
40
50
35
Volume (ccg STP)

45
Volume (ccg-1 STP)

40 30
-1

35 25
30
20
25
15 adsorption
20 adsorption
desorption
desorption 10
15

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Relative pressure (P/P 0) Relative Pressure( P/P 0 )
EMD Tw15 EMD Tw30

Fig. 5—Nitrogen adsorption and desorption isotherms.

low. The effect of varying the concentration of TX-100 in oxidation of Mn2+/3+ reactions. In the case of the non-
the EMD is clearly reflected in the voltammograms. The ionic surfactant, Tw20, the voltammogram given in
EMD produced in the presence of surfactant contains Figure 6(d) does not reveal much influence of Tw in
adsorbed surfactant molecules and thus has a higher terms of the peak intensities relative to EMD0. Although
surface area to that of the bare EMD (undoped), resulting the nature of crystallization is not altered due to the
in higher redox peak currents during the reduction and presence of TX-containing EMD (as evidenced by XRD,

METALLURGICAL AND MATERIALS TRANSACTIONS E

 0.004 0.004 - 0.025 V

A1
0.002 - 0.097 V 0.002

Current / A
Current / A
A1
0.000 0.000

C1
-0.002 C1
-0.002 - 0.390 V
- 0.430 V
-0.004
-0.004 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
Potential vs. Hg/HgO / V
Potential vs. Hg/HgO / V
(c) TX30
(a) EMD0

0.030 V
0.004 0.004

A1

0.002 0.002 - 0.085 V

Current / A
Current / A

A1
0.000
0.000

-0.002 C1
-0.002 C1 - 0.450 V

- 0.450 V -0.004
-0.004 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
Potential vs. Hg/HgO / V
Potential vs. Hg/HgO / V
(d) TW15
(b) TX15

Fig. 6—Cyclic voltammograms of the EMD samples (a) containing no surfactant, (b) Triton (TX15), (c) Triton (TX30), and (d) Tween (Tw15).

Figure 1), the change in morphology and larger surface intensity, corresponding to conductive carbon as an
area (Table I) was reflected in the electrochemistry, additive in the cathode material to improve the conduc-
suggesting a higher rate of nucleation when TX-100 is tivity. In addition to the parent MnO2 diffraction lines,
present in the electrolyte. the evolution of new diffraction peaks corresponding to
The results of cyclic voltammetry for these studies MnOOH and Mn3O4 is seen in Figure 8. The diffraction
were also confirmed by galvanostatic charge/discharge pattern obtained for EMD material prepared in the
studies (Figure 7). As expected, the available discharge presence or absence of surfactants shows the surfactants
capacity for the TX-added EMD showed 273 mAh g1 did not influence the formation of the discharged
(Figure 7(b)), while the surfactant-free EMD showed products. The product obtained is in accordance with
just 170 mAh g1 (Figure 7(a)), additional capacity of the well-known mechanism of c-MnO2 as an insertion
38 pct. The difference in capacities between the various electrode in aqueous KOH electrolyte involves H+
concentrations of TX (Figures 7(b) to (c)) is negligible insertion forming the manganese oxy hydroxides
(only 20 mAh g1). On the other hand, non-ionic (MnOOH) and manganese (III) oxide (Mn3O4).[24–26]
surfactant (Tw) in Figure 7(d) showed only a slight The TEM images of the discharged material, on the
improvement of 181 mAh g1 relative to EMD0. The other hand, show a different but intriguing picture. The
galvanostatic and potentiostatic results of this study TEM images of the cathode (c-MnO2) material prepared
agree well. in the presence of surfactants before and after discharge
To investigate the role of the discharge mechanism in are shown in Figure 9. The TEM image of the cathode
the alkaline KOH electrolyte, the cathode material after material before discharge (Figure 9(a)) showed the
discharge was thoroughly washed with distilled water, parent MnO2 and carbon as additive, but after discharge
and the variation in crystal structure analyzed through (for undoped; Figure 9(b)), the regions A and B
XRD before and after the electrochemical processes. correspond to MnO2 with a small amount of potassium
The XRD results are displayed in Figure 8. The XRD (EDS analysis not shown here). For TX-added EMD,
pattern shows a peak at 2h = 27 deg with higher Figure 9(c), all regions labeled A to E showed MnO2

METALLURGICAL AND MATERIALS TRANSACTIONS E

 1.6 EMD materials exhibit a sharp drop in potential at
1.3 V, although the cells have been tested under identical
conditions. These curves in Figures 7(a) and (d) show
1.2 typical anodic limited behavior, suggestive of passiv-
Cell Potential / V

(a) (b) ation of zinc on the surface of the MnO2 cathode. This
(d)
(c) has been confirmed through the simultaneous PIXE and
0.8 RBS analyses of the EMD material before and after
discharge. A typical bar graph containing several steps
corresponding to elements such as K, Mn, Zn, and C is
0.4
given in Figure 10. The variation in the quantity of
manganese and carbon across the electrodes before and
after discharge appears to be fairly constant. However,
it is quite important to note the differences seen in the
0.0
0 100 200 300
concentrations of the potassium and zinc. For the
-1 electrode before discharge, the concentrations of potas-
Discharge Capacity / mAh g
sium and zinc are nil, as is expected given there is no
potassium and zinc in the fresh sample. For the
Fig. 7—First discharge behavior of the EMD samples (a) containing
no surfactant, (b) Triton (TX15), (c) Triton (TX30), and (d) Tween discharged electrode, the concentration of potassium
(Tw30). ion is increased, particularly in the case of TX-100 as the
surfactant. This reflects the TEM microscopy analysis
and further supports the potassium intercalation. A
C Ο MnO 2 Δ Mn 3 O4 further notable point in the RBS analysis, in Figure 10,
C Carbon • MnO OH is the high concentration of zinc for EMD0 and Tw30
electrodes. This shows that the zinc has been precipi-
• Ο• Ο
Ο Δ tated on the cathode as Zn(OH)2, thus inhibiting further
(d)
• • Ο
electron transfer. Hence, the two cells are anode limited
Ο
•
Ο • with a sharp fall of potential as seen in the discharge
• Ο
Δ Ο curve Figures 7(a) to (d).
(c)
The cumulative discharge capacities of the EMD samples
Ο• Ο in the presence and absence of surfactant TX and Tw are
• ΟΔ • Ο
(b) • shown in Figure 11. The charge and discharge character-
• Ο•
istics enable evaluation of the suitability of the prepared
Ο EMD as a battery material. In Figure 6, it was clearly
Ο
(a)
Δ • • Ο indicated that surfactant works up to an optimum level,
above which an alteration in the cyclic behavior is observed.
Interestingly both surfactants employed in this work
20 30 40 50 60 70 showed their best performance at their 15 t (CMC 9 15)
2 theta / degrees value. EMDTX15 and EMDTw15 showed cumulative dis-
charge capacity of ~2213 and ~2242 mAh g1 at 15th cycle,
Fig. 8—X-ray diffraction patterns of the discharged EMD samples respectively, against a cumulative discharge capacity of
(a) containing no surfactant, (b) Triton (TX15), (c) Triton (TX30), ~1300 mAh g1 for EMD0 at 11th cycles which remain
and (d) Tween (Tw30). constant so far. The discharge capacity of 280 mAh g1 was
obtained from the EMD0 against 170 mAh g1 for the
reported values for EMD in the absence of additives.[52] The
with a larger amount of potassium. For Tw-added initial discharge capacity, above and below which the
EMD, Figure 9(d), except the regions 3 and 4, all discharge capacity decreases, of 270 mAh g1 was observed
regions are identified as K free MnO2 suggesting only for EMDTX15, where as the initial discharge capacity of
a small amount of potassium ion is present. Almost all EMDTw15 was 240 mAh g1, somewhat less than the
the reported literature on MnO2 in alkaline electrolyte EMDTw10 (275 mAh g1). In spite of this, the EMDTX15
describe either protonation[24–27] or lithiation[3,15,28] as showed better cycling stability and higher cumulative
the mechanism involved during discharge but an discharge capacity than EMD0, EMDTw10, and EMDTw30.
unusual observation of potassium ion insertion or as The available cumulative discharge capacity for EMD0 after
surface film formation is evidenced in this study. 10th cycles was quite low of ~1300 mAhg g1, so the cycling
Interestingly, the extent of potassium ion increases with studies were halted well before the cut-off range. The
TX-100 as the surfactant, and this conclusion is further characteristic property of non-ionic surfactant is a tendency
supported by Rutherford backscattering experiments, to accumulate in electrolytic solution that enhances the
discussed in the next section. The observed microscopy adsorption. In the case of TX, nucleation and growth of
results are in parallel with the cyclic voltammograms electrodeposited MnO2 grains in a preferential direction was
and discharge behavior, indicating that the surfactant promoted while favoring a high surface area of the deposit.
facilitates K+ insertion. TX-100 renders the surface more hydrophobic and acts as a
A close inspection of Figures 7(a) and (d) reveals that molecular spacer[21,53] through the steric factor of the phenyl
the discharge curves for surfactant-free and Tw-added ring, while the Tw counterpart is without a phenyl ring. This

METALLURGICAL AND MATERIALS TRANSACTIONS E

Fig. 9—TEM images of EMD (c-MnO2) samples (a) before and (b) after discharge containing no surfactant, (c) discharged Triton (TX30), and
(d) discharged Tween (Tw15).

25 difference in structure observed for non-ionic surfactant led

Relative Composition / at. (%)

K
EMD 0 EMD (TX30 ) EMD (TW30 ) Mn
a role for Mn deposition by facilitating uniform current
20 Zn distribution resulting in Mn-TX complexes. Hence, the
C presence of TX molecules facilitated the superior electro-
15 chemical performance of discharge capacity 275 mAh g1.
The enhancement of cycle life in the presence of the two
surfactants may be attributed to the following: EMDTX15
10
offers high surface area with large pore size (mesopores)
with nanofibrous morphology (as per the FESEM images)
5 which decreases the solid state diffusion path length of
protons and electrons into and out of the bulk of MnO2, as
0 happened in MnO2 films.[54] As explained before, the large
pore size facilitates the proton diffusion inside the MnO2
Before / After Before / After Before / After
discharge discharge discharge electrode, during charge–discharge cycling resulting good
cyclic stability. Second in terms of conductance, EMDTX15
Fig. 10—Simultaneous PIXE and RBS analyses of the possible ele- consists of smaller particles with porous structures, and
ments present in the discharged EMD samples containing no surfac- EMDTw15 composed of spherical nanoparticles (as shown in
tant, Triton (TX30), and (d) Tween (Tw30). FESEM image), which allow better contact with the

METALLURGICAL AND MATERIALS TRANSACTIONS E

 2250
300 e
2000 d
c
Discharge capacity(mAh g-1)

discharge capacity(mAhg )
-1
250 1750
b
1500
200

Cumulative
1250
150
a
1000

100 750
e
d 500
50 c
b 250
a
0 0
2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
Cycle number Cycle number
(a) (b)

2250
300 d
Discharge capacity( mAh g -1)

2000

discharge capacity (mAhg )

c

-1
250 1750
1500 b
200
Cumulative
1250
150 a
1000
750
100 d
500
50
c
250
a b 0
0 0 2 4 6 8 10 12 14 16
0 2 4 6 8 10 12 14 16
Cycle number
Cycle number
(c) (d)

Fig. 11—(a) Discharge capacities versus cycling behavior of (a) EMD0, (b) EMDTX30, (c) EMDTX5, (d) EMDTX10, (e) EMDTX15. (b) Cumulative
discharge capacities versus cycling behavior of (a) EMD0, (b) EMDTX30, (c) EMDTX5, (d) EMDTX10, (e) EMDTX15. (c) Discharge capacities ver-
sus cycling behavior of (a) EMD0, (b) EMDTw30, (c) EMDTw10, (d) EMDTw15. (d) Cumulative discharge capacities versus cycling behavior of
(a) EMD0, (b) EMDTw30, (c) EMDTw10, (d) EMDTw15.

graphite powder. Thermal stability of EMDTX15 and
EMDTw15 is also responsible for the good cyclic behavior
of EMDTX15 and EMDTw15 in comparison to EMD0, which
resists the capacity fading during charge–discharge cycling.
IV.
EMD samples have been prepared from purified
manganese sulfate solution obtained from manganese
cake in the presence or absence of surfactants. The
surfactants played a role in the intercalation behavior. A
unique observation of potassium ion intercalation in
c-MnO2 structure is reported. From the present study,
the following conclusions can be drawn.
1. EMD samples prepared in the presence and absence
of surfactants show the characteristic peak of
c-MnO2 and are electrochemically active. Addition
of surfactant and surfactant type (either TX-100 or
Tw-20) has no significant effect on the crystal struc-
ture of EMD.
2. FESEM images indicate that presence of surfactants
in the electrolytic bath plays a key role in control-
ling the growth and nucleation of the deposited
material and hence modifies the structure accord-
ingly giving porous morphology to the material.
3. The presence of surfactant enhances the electro-
CONCLUSIONS chemical activity of EMD, not only by modifying
the structure but also by increasing the BET surface
area and porosity of the EMD samples. EMD
doped with surfactants has good cycle life in spite
of a decrease in initial discharge capacity, relative
to surfactant-free EMD.
4. The presence of surfactant affects the thermal sta-
bility of the material and also brings mild changes
in spectroscopic analytics. There optimum concen-
tration of surfactant must be essential for the posi-
tive consequences in the performance and behavior
of the EMD.
5. TEM imaging and RBS analysis showed evidence
for potassium intercalation with the extent being
greater for TX-100 as the surfactant. The

METALLURGICAL AND MATERIALS TRANSACTIONS E

 discharged product of the EMD material showed
evidence of protonation.
ACKNOWLEDGMENTS
The authors are thankful to Prof. B.K. Mishra,
Director, CSIR-IMMT, Bhubaneswar, for permitting
to publish the paper. Authors also thank to the Minis-
try of Earth Sciences for partly sponsoring the work.
The first author, A. Biswal is thankful to Council of
Scientific and Industrial Research (CSIR), India for
awarding Senior Research Fellowship. The author
(M.M.) wishes to thank both the Australian Research
Council (ARC) and Australian Nuclear Science and
Engineering (AINSE/AINGRA 08048) for providing
experimental time to enable work on TEM and RBS
facilities.
REFERENCES
1. O. Schilling and J.R. Dahn: J. Appl. Crystallogr., 1998, vol. 31,
pp. 396–406.
2. Y. Chabre and J. Pannetier: Prog. Solid State Chem., 1995, vol. 23,
pp. 1–130.
3. M. Minakshi, P. Singh, T.B. Issa, S. Thurgate, and R. De Marco:
J. Power Sources, 2004, vol. 138, pp. 319–22.
4. D. Balachandran, D. Morgan, G. Ceder, and A. van de Walle: J.
Solid State Chem., 2003, vol. 173, pp. 462–75.
5. A. Urfer, G.A. Lawrance, and D.A.J. Swinkles: J. Appl. Electro-
chem., 1997, vol. 27, pp. 667–72.
6. K. Kordesch and M. Weissenbacher: J. Power Sources, 1994,
vol. 51, pp. 61–78.
7. W. Jantscher, L. Binder, D.A. Fiedler, R. Andreaus, and K.
Kordesch: J. Power Sources, 1999, vol. 79, pp. 9–18.
8. S. Chou, F. Cheng, and J. Chen: J. Power Sources, 2006, vol. 162,
pp. 727–34.
9. K. Kordesch, W. Harer, W. Taucher, and K. Tomantscher: Pro-
ceedings of 22nd Intersociety Energy Conversion Engineering Con-
ference, 1987, vol. 2, p. 1102.
10. H. Malankar, S.S. Umare, K. Singh, and M. Sharma: J. Solid
State Electrochem., 2010, vol. 14, pp. 71–82.
11. A.M.A. Hashem, K.S. Abou-El-Sherbini, S. Zein-El-Abedin, and
H. Abbas: J. Mater. Sci. Technol., 2006, vol. 22, pp. 25–30.
12. A.M.A. Hashem, K.S. Abou-El-Sherbini, M.H. Askar, and A.M.
Hashim: J. Mater. Sci. Technol., 2001, vol. 17, pp. 351–54.
13. A.G. Kholmogorov, A.M. Zhyzhaev, U.S. Kononov, G.A.
Moiseeva, and G.L. Pashkov: Hydrometallurgy, 2000, vol. 56,
pp. 1–11.
14. J.M.M. Paixdo, J.C. Amaral, L.E. Memoria, and L.R. Freitas:
Hydrometallurgy, 1995, vol. 39, pp. 215–22.
15. M. Minakshi, P. Singh, T.B. Issa, S. Thurgate, and R. De Marco:
J. Power Sources, 2004, vol. 130, pp. 254–59.
16. A. Biswal, K. Sanjay, M.K. Ghosh, T. Subbaiah, and B.K.
Mishra: Hydrometallurgy, 2011, vol. 110, pp. 44–49.
17. Suhasini and A.C. Hegde: J. Electroanal. Chem., 2012, vol. 676,
pp. 35–39.
18. Y. Li, H. Xie, J. Wang, and L. Chen: Mater. Lett., 2011, vol. 65,
pp. 403–05.
19. R. Jiang, T. Huang, J. Liu, J.H. Zhuang, and A. Yu: Electrochim.
Acta, 2009, vol. 54, pp. 3047–52.
METALLURGICAL AND MATERIALS TRANSACTIONS E
20. H. Zhang, Y. Wang, C. Liu, and H. Jiang: J. Alloys Compd., 2012,
vol. 517, pp. 1–8.
21. S. Devaraj and N. Munichandraiah: J. Electrochem. Soc., 2007,
vol. 154, pp. A901–A909.
22. M. Ghaemi, L. Khosravi-Fard, and J. Neshati: J. Power Sources,
2005, vol. 141, pp. 340–50.
23. R. Khayat Ghavami, Z. Rafiei, and S.M. Tabatabaei: J. Power
Sources, 2007, vol. 164, pp. 934–46.
24. K. Kordesch, J. Gsellmann, M. Peri, K. Tomantschger, and R.
Chemelli: Electrochim. Acta, 1981, vol. 26, pp. 1495–1504.
25. A. Kozawa and J.F. Yeager: J. Electrochem. Soc., 1965, vol. 112,
pp. 959–63.
26. H.S. Wroblowa and N. Gupta: J. Electraonal. Chem., 1987,
vol. 238, pp. 93–102.
27. Y.F. Yao, N. Gupta, and H.S. Wroblowa: J. Electroanal. Chem.,
1987, vol. 223, pp. 107–17.
28. M. Minakshi, P. Singh, and D.R.G. Mitchell: J. Electrochem. Soc.,
2007, vol. 154, pp. A109–13.
29. A. Biswal, B.C. Tripathy, T. Subbaiah, D. Meyrick, and M.
Minakshi: J Solid State Electrochem., 2013, vol. 17, pp. 1349–56.
30. S. Nijjer, J. Thonstad, and G.M. Haarberg: Electrochim. Acta,
2000, vol. 46, pp. 395–99.
31. W.H. Kao and V.J. Weibel: J. Appl. Electrochem., 1992, vol. 22,
pp. 21–27.
32. F.-Q. Xue, Y.-L. Wang, W.-H. Wang, and X.-D. Wang: Electro-
chim. Acta, 2008, vol. 53, pp. 6636–42.
33. Z. Kozarac, S. Nikolic, I. Ruzic, and B. Cosovic: J. Electroanal.
Chem., 1982, vol. 137, pp. 279–92.
34. I. Felhosi, J. Telegdi, G. Palinkas, and E. Kalman: Electrochim.
Acta, 2002, vol. 47, pp. 2335–40.
35. D.S. Wen and B.X. Wang: Int. J. Heat Mass Transf., 2002, vol. 45,
pp. 1739–47.
36. F.J. Srydlowskl, D.L. Dunmire, E.E. Peck, R.L. Eggers, and W.R.
Matson: Anal. Chem., 1981, vol. 53, pp. 193–96.
37. A.M. Rodriguez, M.A.C. Vilchez, and R.D. Alvarez: J. Colloid
Interface Sci., 1997, vol. 187, pp. 139–47.
38. B. Factor, B. Muegge, S. Workman, E. Bolton, J. Bos, and M.M.
Richter: Anal Chem., 2001, vol. 73, pp. 4621–24.
39. H. Abbas and S.A. Nasser: J. Power Sources, 1996, vol. 58, pp. 15–
21.
40. J.B. Fernandes, B.D. Desai, and V.N.K. Dalal: Electrochim. Acta,
1983, vol. 28, pp. 309–15.
41. M.V. Ananth, S. Pethkar, and K. Dakshinamuthi: J. Power
Sources, 1998, vol. 75, pp. 278–82.
42. P. Ruetschi: J. Electrochem. Soc., 1984, vol. 131, pp. 2737–44.
43. J. Fitzpatrick, L.A.H. Maclean, D.A.J. Swinkels, and F.L. Tye: J.
Appl. Electrochem., 1997, vol. 27, pp. 243–53.
44. M.B. McBride: Soil Sci. Soc. Am., 1987, vol. 51, pp. 1466–72.
45. J.P. Zheng, P.J. Cygan, and T.R. Jow: J. Electrochem. Soc., 1995,
vol. 142, pp. 2699–2703.
46. J.H. Sharp and D.M. Tinsley: J. Therm. Anal., 1971, vol. 3, pp. 43–
48.
47. J.B. Fernandes, B.D. Desai, and V.N.K. Dalal: J. Power Sources,
1985, vol. 16, pp. 1–43.
48. M.I. Zaki, M.A. Hassan, L. Pasupulety, and K. Kumari: Ther-
mochim. Acta, 1997, vol. 303, pp. 171–80.
49. J.O. Besenhard, J. Gurtler, P. Komenda, and A. Paxions: J. Power
Sources, 1987, vol. 20, pp. 253–58.
50. H. Adelkhani and M. Ghaemi: Solid State Ion., 2008, vol. 179,
pp. 2278–83.
51. S.M. Davis, W.L. Bowden, and T.C. Richards: J. Power Sources,
2005, vol. 139, pp. 342–50.
52. M.M. Thackeray: Prog. Solid State Chem., 1997, vol. 25, pp. 1–71.
53. P. Matejka, B. Vickova, J. Vohlidal, P. Pancoska, and V.
Baumruk: J. Phys. Chem., 1992, vol. 96, pp. 1361–66.
54. S.C. Pang, M.A. Anderson, and T.W. Chapman: J. Electrochem.
Soc., 2000, vol. 147, pp. 444–50.