Inelastic neutron scattering studies of the proton dynamics in Bi-doped manganese oxides

Article abstract of DOI:10.1149/1.1394038

The inelastic neutron scattering (INS) spectra from 16 to 4000 cm^-1 are presented for lamellar-like manganese dioxides at 30 K: δ-MnO₂ and chemically modified Bi-δ-MnO₂ at various reduction degrees (Bi-δ-MnO _1.96, Bi-δ-MnO_1.88, Bi-δ-MnO_1.78, Bi-δ-MnO_1.70 and Bi-δ-MnO_1.51). Three different types of protons are distinguished: water-like entities at 540 cm^-1 , oxoniumlike entities at 815 cm^-1, and free protonic entities giving a continuum with constant intensity. The nonreduced samples contain equimolecular mixtures of water-like and oxonium-like entities, and there is no visible modification of the spectra attributable to the Bi entities. As the reduction degree increases, the band intensity at 540 cm^-1 decreases and that at 815 cm^-1 increases. Free protons are virtually unaffected. Even for the most reduced sample, there is no evidence for any feature specific to protons inserted in the [MnO₆] layers. Two processes can be distinguished: the removal of water molecules by the solvent on the one hand and the protonation of water-like entities on the other.

Full text of DOI:10.1149/1.1394038

4184
Journal of The Electrochemical Society, 147 (11) 4184-4188 (2000)
S0013-4651(00)01-079-X CCC: $7.00 © The Electrochemical Society, Inc.
Inelastic Neutron Scattering Studies of the Proton Dynamics in
Bi-Doped Manganese Oxides
F. Fillaux,a C. Cachet,b S. F. Parker,c J. Tomkinson,c A. Quivy,d and L. T. Yub,*,z
aLaboratoire Dynamique, Interactions et Réactivité and bLaboratoire d’Electrochimie et Synthèse Organique,
Centre National de la Recherche Scientifique, 94320 Thiais, France
cISIS Pulsed-Neutron Source, Rutherford Appleton Laboratory, Chilton OX11 0QX, United Kingdom
dCentre d’Etude de Chimie Métalurgique, Centre National de la Recherche Scientifique, 94400 Vitry sur Seine, France
The inelastic neutron scattering (INS) spectra from 16 to 4000 cmϪ1 are presented for lamellar-like manganese dioxides at 30 K:
␦-MnO2 and chemically modified Bi-␦-MnO2 at various reduction degrees (Bi-␦-MnO1.96, Bi-␦-MnO1.88, Bi-␦-MnO1.78
,
Bi-␦-MnO1.70, and Bi-␦-MnO1.51). Three different types of protons are distinguished: water-like entities at 540 cmϪ1, oxonium-
like entities at 815 cmϪ1, and free protonic entities giving a continuum with constant intensity. The nonreduced samples contain
equimolecular mixtures of water-like and oxonium-like entities, and there is no visible modification of the spectra attributable to
the Bi entities. As the reduction degree increases, the band intensity at 540 cmϪ1 decreases and that at 815 cmϪ1 increases. Free
protons are virtually unaffected. Even for the most reduced sample, there is no evidence for any feature specific to protons insert-
ed in the [MnO6] layers. Two processes can be distinguished: the removal of water molecules by the solvent on the one hand and
the protonation of water-like entities on the other.
© 2000 The Electrochemical Society. S0013-4651(00)01-079-X. All rights reserved.
Manuscript submitted January 20, 2000; revised manuscript received July 6, 2000.
A detailed view of protons at various sites in manganese dioxides
are not distinguished. However, bands for protons inserted in differ-
ent sites are partially resolved.4 The observed frequencies were ten-
tatively related to local perturbations induced by the Mn4ϩ vacan-
cies. Band intensities provide information on the chemical affinity of
the various sites.
is of importance for the characterization of the electrochemical
activity of these materials commonly used for electrodes in alkaline
batteries. Inelastic neutron scattering (INS) is a powerful technique
for such studies. Because the proton scattering cross section is very
large compared to those of Mn and O atoms, the MnO2 matrix is vir-
tually transparent and the interpretation of the spectra in terms of
proton dynamics at specific sites is straightforward. The observed
frequencies are determined by local potentials experienced by pro-
tons, and different sites, or chemical bonding, can be distinguish-
ed.1-6 The scattering process is due to nuclear interactions, and cross
sections are independent of the nature of the chemical bonding.
Therefore, band intensities for different protonic species are propor-
tional to their relative amounts. The INS technique is also unique in
showing the recoil of free or very weakly bound protonic entities.1,6
These mobile entities are related to the electronic structure and could
play an important role as charge carriers. Finally, the great penetra-
tion depth of neutrons gives information on protons in the bulk. In
contrast to this, with optical techniques it is difficult to distinguish
signals due to protonated species from the MnO2 matrix and band in-
tensities are not simply proportional to the number of nuclei.
Previous INS studies on manganese dioxides at various reduction
degrees have revealed new aspects of the proton dynamics in mate-
rials obtained by electrochemical (EMD) or chemical (CMD) reduc-
tion. The recoil of free protonic entities with mass of 1 amu was ob-
served.1,6 Unfortunately, the INS technique does not provide any in-
formation on the effective charges and Hϩ, H0, HϪ, or any interme-
diate states cannot be distinguished. Charge compensating protons in
Mn4ϩ vacancies form tetrahedral (Hϩ)4 entities rotating freely at the
center of the oxygen octahedra surrounding the vacancies.2 The
gyration radius of these entities (ca. 0.5 Å) is quite similar to the size
of the missing Mn4ϩ ions (ca. 0.53 Å). Temperature effects from 20
to 300 K suggest that free and charge compensating protons are
related. At low temperature, the free protons are delocalized in the
structural channels. As the temperature increases, these protons enter
the Mn4ϩ vacancies to form the very weakly bound (Hϩ)4 entities.
Protons inserted by chemical reduction of MnO2 with cinnamic alco-
hol are located at the center of oxygen octahedra in the structural
channels and behave as isotropic oscillators with mass of ca. 1 amu.3
Protons inserted in the pyrolusite-like or ramsdellite-like channels
In this present paper we report INS studies of the Bi-dopped
manganese dioxides, also referred to as chemically modified man-
ganese dioxides (CMM). The rechargeability of these materials is
significantly better than that of the CMD or EMD materials. In alka-
line media the cycle life number (CLN) can be more than several
hundreds for the CMM and the depth of discharge (Ddch) can be as
large size 1.7 e/Mn.7-12 As opposed to these performances, the CLN
is only ca. 60 for Ddch ca. 0.55 e/Mn, even for the best EMD or
CMD materials.13 The CLN can be increased to ca. 150-200, but
only for a rather small Ddch of ca. 0.35 e/Mn.14 The CMM materials
have also interesting performances in nonaqueous solvents (CLN >
100 for a Ddch of 0.9 e/Mn)15 comparable to those of the spinel-like
Li2Mn2O4 materials.
X-ray diffraction (XRD) patterns of Bi-dopped manganese diox-
ides, with chemical formulas BiyMnOx, correspond to ill-crystallized
structures,15 compatible with a layer-like structure, such as in chal-
cophanite, ZnMn3O7и3H2O,17,18 consisting of sheets of H2O mole-
cules between layers of edge sharing [MnO6] octahedra. The dis-
tance between two consecutive layers is about 7.17 Å. In chalco-
phanite, Zn atoms are located between the water and the [MnO6] lay-
ers. The water molecules are supposed to form open double hexago-
nal rings. Since manganese atoms occupy only six among seven pos-
sible octahedral sites, there are structural Mn vacancies due to the
Zn ions.
During the discharge process, protons (Hϩ) and electrons (eϪ) are
introduced into the material. The electrons reduce Mn(IV) into
Mn(III) atoms and the insertion of 1 H/Mn could give a structure sim-
ilar to that of feitknechtite.19,20 It is not yet clear whether protons are
inserted between the [MnO6] layers, in which case they should inter-
act with the water molecules, or inside the oxygen octahedra as in the
CMM and CMD materials.3,4 The INS experiments performed on
CMM samples with various insertion (or reduction) degrees present-
ed below were undertaken to determine the location of the inserted
protons and their possible interactions with water molecules and/or
with Bi(III) atoms. It turns out that there is no evidence for protons
inserted in the oxygen octahedral sites. Protons in the CMM material
exist only as water-like and oxonium-like entities with concentrations
depending on the reduction degree.
* Electrochemical Society Active Member.
E-mail: yu@glvt-cnrs.fr
z
Downloaded on 2015-06-21 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (11) 4184-4188 (2000)
4185
S0013-4651(00)01-079-X CCC: $7.00 © The Electrochemical Society, Inc.
is the total amount of sorbed and structure water molecules. The
amount p of OH groups includes surface sites inserted protons. Unfor-
tunately, the amount of water was strongly affected during The ele-
mental analysis, and the results are not conclusive Nevertheless, there
is at least one water molecule per manganese atom. The thermogram
analysis reveals that samples heated at 125, 350, and 450ЊC for 16 h
(Bi-Mn125, Bi-Mn350, and Bi-Mn450, respectively) are free of H2O.
Table I. Redox level and proton insertion degree for the samples
studied with INS.
Sample no.
x in MnOx
rH inserted H (mole/Mn)
Bi-MnO1.96
Bi-MnO1.88
Bi-MnO1.78
Bi-MnO1.70
1.96
1.88
1.78
1.70
1.51
1.98
1.99
0.08
0.24
0.44
0.60
0.98
0.04
0.04
X-ray data.—The XRD patterns were obtained with a Philips
powder diffractometer equipped with a curved graphite monochro-
mator (␭ ϭ 1.7889 Å). The CMM samples (Fig. 1) are compatible
with ill-crystallized layered-like structures analogous to that of chal-
cophanite.
Bi-MnO1.51
␦MnO1.96.a(Bi free)
Sample Bi-Mn450 (H-free)
INS spectra.—The INS spectra were obtained with the TFXA
spectrometer at the ISIS pulsed-neutron source, Rutherford Apple-
ton Laboratory, Chilton, U.K.19 The spectrometer resolution was
⌬␻/␻ ca. 2%. About 10 g of sample were wrapped in aluminum foil
under a dry atmosphere. They were rapidly loaded into a closed
cycle refrigerator. The sample temperature was ca. 30 K. The sam-
ple sizes were 2 ϫ 4 ϫ 0.2 cm, and the total scattered intensity was
less than 5% of the incident neutrons. Under such conditions, multi-
ple scattering is negligible. Spectra were normalized according to
the sample weights in the beam.
Experimental
Sample preparation and characterization.—The Bi-doped man-
ganese dioxide was obtained by slow decomposition of an acidic
solution of KMnO4 in presence of Bi(NO3)3.18 500 mL of 1 M
HNO3 solution containing 0.04 M KMnO4 and 0.04 M Bi(NO3)3
were kept steady at 22ЊC for one week while the CMM precipitates
slowly. Compared to the coprecipitation reaction, which lasts only
for a few hours, this preparation yields a more homogeneous sample.
The precipitate was filtered under vacuum and washed with distilled
water until the washing solution was colorless. The precipitate was
Results
The INS spectrum of the Bi-doped sample with the initial inser-
tion degree of ca. 0.08 H/Mn (Bi0.16MnO1.96 in Fig. 2), reveals a
broad asymmetric band extending itself from 300 to 1500 cmϪ1. The
profile is much broader than bands previously observed for other
MnO2 varieties and assigned to librations of water-like entities
and/or OH groups at the grain surface.1-4 The lattice density-of-
then dried at 70ЊC for 24 h. Chemical titration gives Bi-MnO1.96
.
The layered-like manganese dioxide (␦-MnO1.96) was prepared from
decomposition of a pure 0.04 M KMnO4 solution for one week.
The CMM was reduced at various levels by cinnamic alcohol
(Table I), in the same way as previously reported for ␥-MnO2.3,4 Ele-
mental analysis and chemical redox titration data are compatible with
the chemical formulas: (BiO1.5)yMnOx(OH)p(H2O)m with y < 0.24. m
Figure 1. XRD pattern of the Bi-free and CMM samples at room tempera-
ture. Bi-Mn450 was heated at 450ЊC for 16 h.
Figure 2. INS spectra of the Bi-free ␦-MnO1.96 and CMM samples at 30 K.
Downloaded on 2015-06-21 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

4186
Journal of The Electrochemical Society, 147 (11) 4184-4188 (2000)
S0013-4651(00)01-079-X CCC: $7.00 © The Electrochemical Society, Inc.
states peaks at ϳ200 cmϪ1 with rather modest intensity. Conse-
quently, the proton dynamics is largely independent of the lattice
vibrations. The spectrum of the Bi-free ␦-MnO1.96 sample is very
similar. There is no visible modification to be attributed to Bi enti-
ties. The spectrum of the heat-treated sample (Bi-Mn450 in Fig. 2)
confirms that all protons have been removed and that the contribu-
tion of the MnO2 matrix is negligible.
The INS profile between 300 and 1500 cmϪ1 depends on the in-
sertion degree of the CMM (Fig. 2). The spectra are well represented
with two Gaussian profiles at constant frequencies (540 and
815 cmϪ1, respectively) and a flat continuum (see Fig. 3 and Table II).
For the most reduced sample, the component at 540 cmϪ1 is no longer
visible. The spectral intensity above 900 cmϪ1, which cannot be rep-
resented with a simple Gaussian profile, is attributed to traces of cin-
namic alcohol (see Table I).
In the plot of the band intensities as a function of the proton in-
sertion level y (MnO2Hy Fig. 4) equivalent to the reduction degree x
(MnOx, with y ϭ 2 Ϫ x), two domains can be distinguished. For y <
0.22 (x > 1.88) the intensities of the two components decrease simul-
taneously. Then, as y increases the intensity of the band at 540 cmϪ1
decreases while the band intensity at 815 cmϪ1 increases. In this
range, the intensity for each component is virtually proportional to
y, and the total intensity for the two components is virtually a con-
stant (see Table II).
Underneath the broad bands there is a continuum of intensity
extending over the whole spectrum with virtually constant intensity
(Fig. 3). This continuum is due to the recoil of free protonic entities
with mass of ca. 1 amu.1,6 A straightforward comparison with the
INS spectrum of the EMD material4 reveals that there are about
twice as many recoiling entities in the CMM. However, the missing
manganese atom in the CMM structure (ca. 1/10) is compensated by
Bi ions. Therefore, the recoiling protons must be related to structur-
al defects, rather than Mn4ϩ vacancies. This is confirmed by the
great similarity of the INS spectra of the CMM and Bi free ␦-MnO2
analogue (Fig. 2).
Most of the INS intensity is removed by heat-treatments even
after heating at the rather modest temperature of 125ЊC. This is in
accord with the thermal analysis.15 This rather low binding energy
for protonic entities in the CMM contrasts to previous observation
on the EMD.1
Discussion
The band at 540 cmϪ1 is attributed to librations of water-like
entities, located in between the [MnO6] layers or at the grain surface.
The surface OH groups are also likely to appear in this frequency
range. These water-like entities disappear as the reduction degree
increases.
The band centered at 815 cmϪ1 is tentatively attributed to oxoni-
um entities. It is analogous to the broad band with several maxima
previously observed in hydrated ␤-alumina in the same frequency
range20 and attributed to a manifold of partially resolved librational
modes of protonated water molecules such as oxonium entities H3Oϩ,
or hydrated analogues. The similar mean frequencies and bandwidths
suggest identical interaction between protonated species for the two
systems. Presumably, the fine structure is hidden by disorder in the
CMM material.
Because the Bi ions have no visible contribution on the spectra,
there is no need to consider explicitly these entities in the following
discussion. Consequently, the nonreduced CMM can be schematical-
ly represented as
MnO2Ϫn(OH)2n(H2O)m
[1]
The OH groups are supposed to be at the grain boundaries while
the water molecules are either in between the [MnO6] layers (struc-
tural water) or physically sorbed. According to the strong acid char-
acter of surface groups emphasized by Kozawa,21 INS spectra sug-
gest that these entities dissociate in the solid phase as
Figure 3. Band decomposition into Gaussian components for the INS spec-
tra of the CMM samples at 30 K for various reduction degrees.
2nϪ
[MnO2Ϫn(OH)2n(H2O)m] } [MnO
, (H2O)mϪ2n, (H3Oϩ)2n] [2]
2ϩn
Downloaded on 2015-06-21 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (11) 4184-4188 (2000)
4187
S0013-4651(00)01-079-X CCC: $7.00 © The Electrochemical Society, Inc.
For further reduction degrees, the total amount of protons in the
material is apparently independent of the insertion level. This para-
doxical result suggests that the chemical reduction process must be
decomposed into two elementary mechanisms. First, protonation of
water-like molecules as a direct consequence of the chemical reduc-
tion. Second, partial dehydration by the solvent, which removed the
same amount of protons from the sample. Therefore
Table II. Band decomposition into Gaussian components of the
INS spectra.
Sample
Continuum Frequency Width Surface area
(height in a.u.) (cmϪ1
)
(cmϪ1
)
(a.u.)
␦-MnO2
1.19
1.32
1.91
1.47
1.12
1.00
507
124
340
[MnO2/3nϪ
2ϩ2/3n, (H2O)4/3n, (H3Oϩ)4/3n] ϩ xHϩ ϩ xeϪ
854
792
486
485
497
}
851
517
290
134
514
305
f
[MnO(2/3nϩx Ϫx/3)Ϫ, (H2O)4/3nϪx, (H3Oϩ)4/3nϩ2/3x] ϩ x/2(H2)D [4]
Bi-␦-MnO1.96
Bi-␦-MnO1.88
Bi-␦-MnO1.78
Bi-␦-MnO1.70
Bi-␦-MnO1.51
2ϩ2/3nϪx/6
}
825
543
281
135
487
161
Since there is no observable water-like entities in the most re-
duced sample, 4/3n ϳ 1 and the chemical reaction can be rewritten
in a much simpler form as
}
815
543
268
135
325
123
[MnOϪ2.5, H2O, H3Oϩ] ϩ xHϩ ϩ xeϪ
}
f
815
543
268
135
362
066
(1ϩ2/3x)Ϫ
[MnO
, (H2O)1Ϫx, (H3Oϩ)1ϩ2/3x] ϩ x/2(H2O)D [5]
2.5Ϫx/6
}
If water-like protons and oxygen atoms are removed from grain
boundaries during the reduction process the grain mean-size could
increase and contribute to changes in the diffraction patterns.
Finally, the reduction process has little consequence on the con-
tinuum due to free protonic entities. The intensity can be regarded as
a constant with fluctuation of ca. Ϯ20% not clearly correlated to the
reduction degree (see Table II). Presumably, free protons are more
closely related to the structure (e.g., defects and vacancies) rather
than to the reduction degree.
815
268
431
—
—
—
815
268
457
The band intensity ratio for the nonreduced and Bi-free samples
(I540/I815 ϳ 0.7, see Fig. 4 and Table II) is compatible with an equi-
molar mixture of H2O and H3Oϩ entities [m ca. 4n in formula 2
gives MnO22nϩϪn, (H2O)2n, (H3Oϩ)2n].
For MnO1.88 there is a large decrease of the amounts of H2O and
H3Oϩ entities, compared to the nontreated sample. This change is
quite different from the smooth variations with opposite signs ob-
served for the more reduced samples. We suppose this is due to
dehydration by the solvent, independent of the reduction process.
Compared to the nontreated sample, band intensities for MnO1.88
are multiplied by factors of ca. 0.5 and ca. 0.6, respectively. Linear
extrapolation for x ϭ 0 (Fig. 4) suggests that about one-third of
water-like and oxonium-like entities is removed by the solvent.
Therefore, the dehydration process of the solid phase in contact with
the solvent can be then represented as
Conclusion
The INS spectra reveal three different types of protons in the
lamellar manganese dioxides, either Bi doped or not. Water-like pro-
tons including structural water, surface OH groups, and/or water mol-
ecules physically sorbed at the grain boundaries are observed at
540 cmϪ1. Oxonium-like entities, on the other hand, give a broad
band centered at 815 cmϪ1. Underneath these bands, a continuum
with constant intensity over the whole spectral range is attributed to
the recoil of free protons, presumably associated to structural defects.
According to band intensities, the original sample corresponds
approximately to [MnO1.5Ϫ
2.75 , (H2O)1.5, (H3Oϩ)1.5]. These very large
[MnO22nϩϪn, (H2O)2n, (H3Oϩ)2n]
amounts of water and oxonium-like entities are quite at variance
from those anticipated for the chalcophanite-like structure which
contains only about one water-like entity per Mn atom. It transpires
that the CMM material must be regarded as rather small clusters of
hydrated MnO2 entities corresponding to a very strong acidic medi-
um containing an equimolar mixture of water-like and oxonium-like
entities. The INS spectra of the Bi-free and Bi-containing ␦-MnO2
are almost identical and should correspond to very similar structures.
Unfortunately, the water content estimated from elemental analysis
is corrupted by large errors and cannot be compared with those de-
rived from the spectra.
4/3nϪ
r MnO
, (H2O)4/3n, (H3Oϩ)4/3n] ϩ 5/3nH2OD [3]
2ϩ2/3n
During the chemical treatment, dehydration by the solvent, re-
duction of the manganese atoms, and protonation of water-like en-
tities occur simultaneously. It is possible to distinguish a specific
dehydration effect of the solvent independent of the reduction pro-
cess. This corresponds to the removal of ca. 1.25 water-like entities
per Mn atom and the solid phase in contact with the solvent can be
represented as: [MnOϪ2.5, H2O, H3Oϩ] or [MnOϪ2.5, H5Oϩ2 ]. Reduc-
tion and proton insertion occur without any visible change of the
total amount of protons. We suppose that about one-third of the oxy-
gen atoms released by the reduction of manganese atoms neutralize
oxonium entities: H3Oϩ ϩ 1/2 O2Ϫ r 3/2H2O. These water mole-
cules are removed from the solid phase by the solvent. Presumably,
significant changes of the chemical bonds and local structures
should occur. Finally, the most reduced sample can be represented
as: [MnO52/3.5ϪϪ1/6, (H3Oϩ)5/3]. In contrast to previous observations
on EMD and CMD protons inserted into the [MnO6] octahedra are
not observed.
Figure 4. Variation of the band areas with the H insertion degree. Upper cir-
cles: band at 815 cmϪ1. Lower squares: band at 540 cmϪ1. The solid sym-
bols correspond to the Bi-free ␦-MnO2.
Downloaded on 2015-06-21 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

4188
Journal of The Electrochemical Society, 147 (11) 4184-4188 (2000)
S0013-4651(00)01-079-X CCC: $7.00 © The Electrochemical Society, Inc.
10. D. Y. Qu, B. E. Conway, L. Bai, Y. H. Zhou, and W. A. Adams, J. Appl. Elec-
The continuum of intensity due to free protons is little affected by
trochem., 23, 693 (1993).
11. L. Bai, D. Y. Qu, B. E. Conway, Y. H. Zhou, and W. A. Adams, J. Appl. Elec-
trochem., 140, 884 (1993).
12. H. Ouboumour, C. Cachet, M. Bodé, J. L. Ginoux, M. Laborde, C. Lévy-Clément,
and L. T. Yu, Prog. Batteries Battery Mater., 13, 170 (1994).
13. K. Kordesch, J. Gsellmann, M. Peri, K. Tomantschger, and R. Chemelli, Elec-
trochim. Acta, 26, 1495 (1981).
14. W. Harer and K. Kordesch, in Proceedings of the Third MnO2 Symposium,Vol. 2,
p. 231, Graz (1985).
15. S. Bach, J. Pereira-Ramos, M. Bodé, C. Cachet, and L. T. Yu, Electrochim. Acta,
40, 785 (1995).
16. R. G. Burns and V. M. Burns, in Proceedings of Manganese Dioxide Symposium,
Vol. 1, A. Kozawa and R. J. Brodd, Editors, p. 306, Published by the IC Sample
Office, Cleveland, OH (1975).
the reduction process. These protons are specific to the crystal struc-
ture rather than to the redox state. This work emphasizes the great
variety of mechanisms contributing to the electrochemical activity of
materials based on manganese dioxides.
References
1. F. Fillaux, H. Ouboumour, J. Tomkinson, and L. T. Yu, Chem. Phys., 149, 459 (1991).
2. F. Fillaux, H. Ouboumour, C. Cachet, J. Tomkinson, G. J. Kearley, and L. T. Yu,
Chem. Phys., 164, 311 (1992).
3. F. Fillaux, C. Cachet, H. Ouboumour, J. Tomkinson, C. Lévy-Clément, and L. T.
Yu, J. Electrochem. Soc., 140, 585 (1993).
4. F. Fillaux, C. Cachet, H. Ouboumour, J. Tomkinson, C. Lévy-Clément, and L. T.
Yu, J. Electrochem. Soc., 140, 592 (1993).
17. A. D. Wadsley, Acta Crystallogr., 6, 433 (1953).
18. M. Bodé, C. Cachet, and L. T. Yu, Prog. Batteries Battery Mater., 14, 31 (1996).
(1995).
19. S. T. Robertson, Computer Analysis of TFXA Data, Neutron Science Div., Ruther-
ford Appleton Laboratory, Chilton, U.K. (1988).
20. Ph. Colomban, F. Fillaux, J. Tomkinson, and G. J. Kearley, Solid State Ionics, 77,
161 (1995).
21. A. Kozawa, J. Electrochem. Soc., 106, 552 (1959).
5. F. Fillaux, C. Cachet, H. Ouboumour, J. Tomkinson, C. Lévy-Clément, and L. T.
Yu, Prog. Batteries Battery Mater., 12, 32 (1994).
6. F. Fillaux, J. Tomkinson, S. M. Bennington, and L. T. Yu, Chem. Phys., 209, 111
(1996).
7. Y. F. Yao, N. Gupta, and H. S. Wroblowa, J. Electroanal. Chem., 223, 107 (1987).
8. H. S. Wroblowa and N. Gupta, J. Electroanal. Chem., 238, 93 (1987).
9. M. A. Dzieciuch, N. Gupta, and H. S. Wroblows, J. Electrochem. Soc., 135, 2415
(1988).
Downloaded on 2015-06-21 to IP 130.194.20.173 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).