Beneficial effects of Mo on the electrochemical properties of tin as an anode material for lithium batteries

Article abstract of DOI:10.1016/j.electacta.2005.10.002

A simple, novel method for improving the electrochemical response of Sn in lithium cells is proposed that involves preparing Sn by a reduction procedure in the presence of Mo powders. Four different Mo_xSn_{1 - x} mixtures (0 < x < 0.26) were electrochemically tested and their structural and textural properties determined by using X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and scanning electron microscopy (SEM). The electrochemical properties of the resulting composites in lithium cells were studied by galvanostatic, step potential electrochemical spectroscopy (SPES) and electrochemical impedance spectroscopy (EIS) measurements. The mixtures were found to consist of crystalline Sn and Mo; however, the presence of the latter element modified the Sn habit in two ways, namely, by significantly decreasing particle size and increasing the reactivity towards oxygen. Although Mo is inert towards lithium, it increased both the discharge capacity and the capacity retention of the electrode in relation to pure Sn. The improved interparticle connectivity, reduced electrolyte decomposition and decreased charge-transfer resistance observed in the Mo-containing samples appear to be beneficial effects of the addition of Mo.

Full text of DOI:10.1016/j.electacta.2005.10.002

Electrochimica Acta 51 (2006) 3391–3398
Beneficial effects of Mo on the electrochemical properties of
tin as an anode material for lithium batteries
a
b
a
b
a,∗
´
´
´
A. Fernandez , F. Martın , J. Morales , J.R. Ramos-Barrado , L. Sanchez
a Departamento de Qu´ımica Inorga´nica e Ingenier´ıa Qu´ımica, Facultad de Ciencias, Campus de Rabanales, Edificio Marie Curie,
Universidad de Co´rdoba, 14071 Co´rdoba, Spain
b Laboratorio de Materiales y Superficies (Unidad asociada al CSIC), Facultad de Ciencias, Campus de Teatinos, Universidad de Ma´laga, Ma´laga, Spain
Received 22 June 2005; received in revised form 12 September 2005; accepted 1 October 2005
Available online 4 November 2005
Abstract
A simple, novel method for improving the electrochemical response of Sn in lithium cells is proposed that involves preparing Sn by a reduction
procedure in the presence of Mo powders. Four different MoxSn1 − x mixtures (0 < x < 0.26) were electrochemically tested and their structural and
textural properties determined by using X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and scanning electron microscopy
(SEM). The electrochemical properties of the resulting composites in lithium cells were studied by galvanostatic, step potential electrochemical
spectroscopy (SPES) and electrochemical impedance spectroscopy (EIS) measurements. The mixtures were found to consist of crystalline Sn
and Mo; however, the presence of the latter element modified the Sn habit in two ways, namely, by significantly decreasing particle size and
increasing the reactivity towards oxygen. Although Mo is inert towards lithium, it increased both the discharge capacity and the capacity retention
of the electrode in relation to pure Sn. The improved interparticle connectivity, reduced electrolyte decomposition and decreased charge-transfer
resistance observed in the Mo-containing samples appear to be beneficial effects of the addition of Mo.
© 2005 Elsevier Ltd. All rights reserved.
Keywords: Anode; Lithium batteries; Tin; Molybdenum
1. Introduction
with variable Li concentrations, cause inhomogeneous volume
expansions that result in anode cracking or pulverization and
lead to a loss of electrical contact or poor conductivity that
detracts from long-term cycling performance [4].
Tin-based compounds have been extensively explored in
recent years as alternative negative electrode materials to the tra-
ditional choice (graphite) in the search for higher energy Li-ion
batteries. Research in this context has focused on the reaction of
tin with lithium [1], which forms various LixSn reversible alloys
with a theoretical capacity ashighas993 Ah kg−1. However, two
key issues hindering their commercialization remain unsolved.
One is related with the capacity loss observed when starting
from tin oxides: the reaction SnOx + 2xLi+ → Sn + xLi2O is
irreversible [2]. The other is the inadequate capacity retention
of lithium alloys that results from the large volume changes
occurring during the formation of different Li–Sn phases.
The formation of LixSn alloys involves up to five different
crystallographic phases spanning the range 0 < x < 4.4 and
volume changes as large as 260% [3]. The coexisting phases,
Volume changes during Li insertion/extraction can be con-
trolled in various ways. The methods most widely used for this
purpose involve active materials of small particle size [5,6],
the use of intermetallic alloys and/or composite host materials
instead of pure metal [7,8], and that of anodes bonded by con-
ductive polymer or a carbon matrix [9,10]. In general, the incor-
poration of tin particles into an inactive matrix with a view to
hindering Li–Sn phase transitions is an effective method. How-
ever, the electrodes thus obtained exhibit capacity values lower
than those of pure Sn. In general, the Sn content in carbona-
ceous materials (e.g. C0.8Sn0.2 [11], C0.42Sn0.58 [12], C0.5Sn0.5
[13]) or MxSn1−x alloys (e.g. Cu6Sn5 [14], Sn2Fe:SnFe3C [15],
Mg2Sn [16], Ni3Sn2 [17]) does not exceed 50%.
In this context, it would be of interest to develop effective
procedures for controlling the Sn/inactive matrix ratio in order
to increase the Sn content in the end product as much as possible.
In this work we used a scarcely explored approach involving the
∗
Corresponding author. Tel.: +34 957 218620; fax: +34 957 218621.
E-mail address: luis-sanchez@uco.es (L. Sa´nchez).
0013-4686/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2005.10.002

3392
A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
dispersion of Sn particles between metallic particles which are
inactive to the electrochemical process and useful to improve the
electrical conductivity of the active material [18,19]. The metal
used should to be inert towards lithium. In this way, one can
obtain Mx–Sn1−x mixtures with a minimal presence of the inert
phase, M. Recently, we found Mo to have a beneficial effect on
the electrochemical properties of Sn1−xMoxO2 solid solutions in
lithium cells [20,21]; however, these mixed oxides have a draw-
back, an irreversible capacity owing to the irreversible uptake of
lithium to a zero valence sate of the elements. In this work, we
used a different strategy to exploit the favorable effect of Mo and
avoid such a restriction; thus, the element was added during the
formation of Sn via a reductive pathway. The electrochemical
response of the Sn electrode was thus clearly enhanced.
Electrochemical measurements were made in swagelok-type
cells, using lithium as the counter and reference electrode. The
electrolyte was Merck battery electrolyte LP 40 (EC:DEC = 1:1,
w/w, 1 M LiPF6). Electrode pellets were prepared by press-
ing, in a stainless steel grid, ca. 4 mg of active material with
polytetrafluoroethylene (PTFE) (5 wt.%), and acetylene black
(10 wt.%). Galvanostatic tests were conducted under a C/3 gal-
vanostatic regime (C being defined as 1 Li+ ion exchanged in
1 h). Step potential curves were recorded at 2.5 mV/0.22 h per
step. Both types of electrochemical measurements were con-
trolled via a MacPile II potentiostat–galvanostat. Impedance
measurements from 0.01 to 25 kHz (amplitude 3 mV) were made
with a Solartron 1470 battery test unit coupled with a Schlum-
berger SI 1255 frequency response analyzer.
2. Experimental
3. Results and discussion
All chemicals were analytical grade. Sn was prepared by
reducing SnCl4·5H2O aqueous solution with KBH4 in the pres-
ence of Mo powder. A large excess of KBH4 solution, which
was adjusted to pH 12 with 6 M KOH, was employed. The
reaction was carried out under continuous stirring to ensure
homogenization of the emulsion. The product was collected
by centrifugation, washed with distilled water and ethanol, and
dried at 150 ◦C during 1 h under vacuum.
Chemical analyses of the samples were performed by atomic
absorption spectroscopy (AAS) and energy-dispersive X-ray
analysis (EDX). X-ray powder diffraction (XRD) patterns were
recorded in steps of 0.02◦ and 1.2 s on a Siemens D5000
X-ray diffractometer, using Cu K␣ radiation and a graphite
monochromator. Scanning electron microscopy (SEM) images
were obtained on a Jeol JMS-6400 microscope.
X-ray photoelectron spectra were recorded on a Physical
Electronics PHI 5700 spectrometer using non-monochromated
Mg K␣ radiation (hν = 1253.6 eV) and a hemispherical anal-
yser operating at constant pass energy of 29.35 eV. The samples
were prepared as pellets 7 mm in diameter. High resolution
spectra were recorded at 45◦ takeoff angle by a concentric hemi-
spherical energy electron analyser operating in the constant pass
energymodeat29.35 eV, usinga720 ␮m-diameteranalysisarea.
The multiregion spectra were recorded in three different areas
separated by 1 mm. The energy scale of the spectrometer was
calibrated by using the Cu 2p3/2, Ag 3d5/2 and Au 4f7/2 photo-
electron lines at 932.7, 368.3 and 84.0 eV, respectively. Binding
energies were corrected by using that for C 1 s of adventitious
carbon (and the methyl group) fixed at 284.8 eV. Survey spec-
tra over the range 0–1200 eV were recorded at a 187.85 pass
energy, each region being scanned several times to ensure an
adequate signal-to-noise ratio. A 3 × 3 mm sample area was
sputtered with 4 keV Ar+; the sputter rate was assumed to be
∼6 nm min−1 as determined for Ta2O5 under identical sputter-
ing conditions. Spectra were processed by using PHI-Access V.6
and Multipak software, both from Physical Electronics. Surface
atomic concentrations were determined from peak areas, using
Shirley background subtraction and sensitivity factors provided
by the spectrometer manufacturer (Physical Electronics, Eden
Prairie, MN).
Four MoxSn1−x mixtures named A–D of compositions over
the range 0 ≤ x ≤ 0.26 (where x is referred to atomic values) were
studied. Their XRD patterns (Fig. 1) were indexed on the Sn or
Mo structure. No peaks belonging to SnOx and/or MoOx phases
wereobserved. TheunitcelldimensionsandcrystallitesizeofSn
– the latter of which was determined from the Scherrer equation
[22] – are shown in Table 1 together with the mixture composi-
tion as determined by chemical analysis. The presence of Mo had
little effect on the structural parameters of Sn obtained except
for a slight decrease in crystallite size that was less marked for
sample C. The two phases apparent and the absence of changes
in the lattice parameters rule out the formation of Mo–Sn solid
solutions.
Judging from the SEM images (Fig. 2), the presence of Mo
altered the morphology of the tin particles. Thus, sample A con-
sisted of particles of ill-defined shape that were 1–3 ␮m in size.
Mo was present as round particles and the Sn particles tended
to form agglomerates around the Mo particles. Moreover, the
presence of Mo strongly decreased the size of Sn particles to a
few tenths of a micron. A uniform distribution of both types of
metallic particles was inferred from the EDX results.
Fig. 1. XRD patterns for the Sn–Mo samples. Mo reflections are marked with
an asterisk.

A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
3393
Table 1
Unit cell dimensions and cristallite size of Sn and composition of the MoxSn1−x mixtures
3
˚
˚
˚
˚
˚
Sample
Composition
a = b (A)
c (A)
V (A )
L (A) 2 0 0
L (A) 1 0 1
A
B
C
D
Sn
5.836 (2)
5.830 (2)
5.826 (4)
5.828 (3)
3.185 (2)
3.185 (3)
3.175 (5)
3.181 (4)
18.587
18.568
18.497
18.538
940
808
920
801
964
826
923
859
Mo0.12Sn0.88
Mo0.17Sn0.83
Mo0.26Sn0.74
The XPS spectra for the samples are shown in Figs. 3 and 4.
The Sn 3d5/2,3/2 doublet for sample A (Fig. 3a) was split into
two components, which suggest the presence of two different
oxidation states of Sn at surface level [23,24]. The low-energy
component was assigned to Sn and the high-energy component
to a SnOx layer coating the particles. Sputtering with Ar+-ion
decreased the intensity of this component and increased that of
the low-energy component. The XPS concentration depth pro-
files for tin and oxygen, Fig. 3b, reveal the disappearance of
oxygen after 6 min of sputtering. On the assumption of a depth
6 nm per min of ion Ar+ etching, a thickness of 36 nm was esti-
mated for the SnOx layer, equivalent to 1.2–3% of the particle
volume. Such a small amount of SnOx phase was undetectable in
the corresponding XRD patterns. Because the estimated thick-
ness was greater than the penetration depth of the X-ray beam,
the observation of two types of Sn (unoxidized and oxidized)
raises some doubts concerning the actual surface composition
of the particles. A model based on the Sn particles uniformly
coated by a SnOx layer of the above thickness would be inconsis-
tent with the XPS data. These are better interpreted by assuming
either the Sn particles to be non-uniformly coated by the oxi-
dized phase or the presence of a fraction of unoxidized particles.
The Sn 3d spectra for the mixtures were similar profile to
those described above (Fig. 4a); however, the intensity of the
high-energy component was now greater than that of the low-
energy component. This suggests an increased oxygen content in
the Mo-containing samples by virtue of their decreased particle
size (45 versus 25% of the oxygen present in pure Sn). The Mo
3d signal-to-noise ratio was very low and increased with etching
Fig. 3. (a) Sn 3d XPS spectra obtained for sample A after (—) 1 min and (- - -)
5 min of argon etching. (b) Variation of the Sn and O atomic percentages as a
function of the Ar+ ion sputtering time.
Fig. 2. SEM images of sample A (a) and C (b).

3394
A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
trum (Fig. 4a) was specially significant even after etching; this
suggests that most part of the oxygen is bound to Sn as the Mo
3d doublet was rather symmetric (Fig. 4b). On grounds simi-
lar to those used above, the thickness of the SnO coating layer
must exceed 240 nm, which is greater than the Sn particle size.
The presence of a small amount of uniformly dispersed amor-
phous SnOx may be used as an alternative model to account for
the presence of such residual oxygen and is consistent with the
XRD patterns, where no traces of crystalline tin oxides were
detected.
The electrochemical response of the samples was measured
in Li/LiPF6 EC-DEC/MoxSn1−x cells. The open circuit poten-
tial of the cell was about 2.5 V a typical value for metallic Sn
versus Li. An abrupt potential drop was observed during the
first discharge (Fig. 5) that was followed by a potential plateau
at 1.4–1.6 V the length of which decreased with increasing pres-
ence of Mo. In fact, no plateau was observed for sample D.
The plateau was previously observed by other groups [25–27]
and has been ascribed to severely irreversible decomposition
of the electrolyte catalyzed by crystal faces of pure Sn [25].
This process only takes place at high potentials (1.5 V versus
Li), where the Sn electrode will contain negligible amounts of
lithium. The electrolyte decomposes until a thick layer is formed
that blocks the catalytic process and hinders further decompo-
sition. Under these conditions, the presence of Mo reduces the
extent of this reaction, which does not occur above an atomic
content x ≈ 0.26. The resulting reduced particle size and also
reduced crystal face size, together with the increased thickness
of the oxygen layer coating the crystal faces, may be the ori-
gin of this effect. Between 1 and 0.0 V, the potential decrease
is smoother and occurs in several steps that correspond to the
formation of various LixSn alloys [28]. In some reports [29],
a plateau at around 0.9 V has been assigned to the reaction of
SnO with Li. Although such a reaction cannot be discarded in
our samples, the fact that the plateau width was hardly altered by
the presence of Mo – which is in contrast with the increased oxy-
gen content found – supports the formation of an alloy. Based
on XRD patterns not shown here, Mo is inert towards lithium;
therefore, the Faradaic yield of Fig. 5 is referred to the amount
of Sn. The amount of lithium inserted below 1 V ranged from
2.8 for pure Sn to 3.2–3.4 for Mo-containing samples (these
values are referred to one Sn atom). Therefore, the presence of
Mo increases the reactivity of Sn towards lithium, probably as a
result of the combined effects of (i) a decreased particle size and
(ii) improved Li diffusion resulting from inhibition of the for-
mation of a decomposed electrolyte layer. The presence of Mo
also improved the reversibility of the Li reaction, as reflected in
the first charge values in Fig. 5 recorded to a cut-off voltage of
1 V. In the Mo-containing samples, the amount of Li removed
was greater than that measured for pure Sn (2.8 versus 1.8 Li
atoms per Sn atom).
Fig. 4. (a) Sn 3d XPS and (b) Mo 3d spectra obtained for sample D after (—)
1 and (- - -) 16 min of argon etching. (c) Variation of the Sn, Mo and O atomic
percentages as a function of the Ar+ ion sputtering time.
(Fig. 4b). Also, its binding energy decreased with etching, which
suggests that some oxygen was bound to Mo. Fig. 4c shows the
depth profile analysis for sample D. The main difference from
sample A is the presence of oxygen in appreciable amounts (ca.
10%) even after long periods of sputtering (>40 min). Both Mo
and, especially, Sn increased in content with etching time; also
the Mo/Sn ratio ranged from 0.3 to 0.4, which is fairly consis-
tent with the bulk composition obtained by chemical procedures.
The contribution of the high-energy component to the Sn spec-
The beneficial effect of Mo was also observed on further
cycling. Fig. 6 shows the variation of the specific capacity as
referred to the total weight of the electrode with the number
of cycles for the Li/LiPF6 EC-DEC/MoxSn1−x cells. As can
be seen, the presence of Mo clearly improved the reversibility
of the electrochemical reaction. Although the four electrodes

A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
3395
Fig. 5. First charge–discharge galvanostatic curves for Li/LiPF6 EC-DEC/MoxSn1−x cells.
tested exhibited a capacity loss on cycling, capacity retention
was higher in the Mo-containing samples. This parameter, R40,
defined as the ratio between the discharge capacity in the forti-
eth and second cycle, ranged from 0.1 for sample A to 0.4–0.6
in the Mo-containing samples. The capacity delivered by the
samples with the highest Mo content tended to level off at ca.
275 Ah kg−1 at the cycling tests. Although this value is lower
than that exhibited by commercial graphite (372 Ah kg−1), the
increased density of the MoxSn1−x resulted in higher volu-
metric capacities (2215–2143 Ah cm−3 versus 840 Ah cm−3 for
graphite). Thecapacitydeliveredbythesemixturesishigherthan
that of Mo1−xSnx solid solutions of composition 0.02 < x < 0.5
prepared as thin films by magnetron sputtering, with a deliv-
ered capacity of 120 Ah kg−1 [30]. However, when these alloys
Fig. 7. Step potential curves for Li/sample A and Li/sample D cells cycled over
the 1.0–0.0 V range.
Fig. 6. Variation of the specific capacity with number of cycles for the Li/LiPF6
EC-DEC/MoxSn1−x cells.

3396
A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
Fig. 8. Ex situ SEM images of the electrodes made from samples A and D, both discharged (a and b) and charged (c an d) at 0.0 and 1.0 V, respectively.
are prepared in nanometric size, they contain a small amount of
oxygen (the nominal composition is Mo1−xSnxO0.2). These two
features lead to a significantly increased capacity (300 Ah kg−1).
These latter results are consistent with those found in this work
and confirm the prominent role played by the particle size and
the presence of residual oxygen.
as in Fig. 2, which suggests that they were coated by LixSn
alloy. Small cracks were also observed in the surface (images
not shown) that became more apparent as the electrodes
were charged up to 1.0 V (Fig. 8c and d) through shrinkage
of LixSn particles during the extraction of lithium [2]. The
expansion/shrinking undergone by the Sn structure during the
alloying/dealloying process has been cited as the origin of the
rapid capacity loss of this electrode and may also be the origin
of the cracks. The dilution effect of the Mo particles may hinder
it. Moreover, these cracks decrease electronic conductivity
between particles aggregates. One interesting observation from
the SEM image of the charged Mo0.26Sn0.74 electrode was the
presence of round particles, assigned to Mo, inside the cracks
(Fig. 8d). This might mitigate the conductivity loss at the cracks
and reduce the electrode impedance. All these features are
bound to improve electrode performance, as illustrated in Fig. 6
for the Mo-containing samples.
This last assumption was confirmed by impedance measure-
ments for Sn (sample A) and Mo0.26Sn0.74 (sample D) elec-
trodes. Fig. 9a shows the impedance spectra for the electrodes
at the fifth cycle and at the charged state (cut-off voltage 1.0 V).
The shape of the two spectra is highly similar: a well-defined
semicircle in the high-frequency domain followed by an incom-
plete semicircle that overlaps with a line in the low-frequency
region. The spectra are typical of the formation of resistive sur-
face films on electrodes. The high-frequency semicircle has been
ascribed to Li migration through surface layers and that in the
mid-frequency region to interfacial charge-transfer [29,33–35].
Thelinearpart of thespectrareflectsthediffusion ofLi+ ions into
In order to shed additional light on the effect of the pres-
ence of Mo on the electrochemical activity of Sn, supplemen-
tary data were obtained by combining step potential electro-
chemical spectroscopy (SPES) with ex situ SEM observations
and impedance measurements. Fig. 7 shows the step potential
curves for Li/sample A and Li/sample D cells cycled over the
1.0–0.0 V range. The different peaks observed during the reduc-
tion and oxidation processes were ascribed to the reversible
formation of LixSn (0.0 < x < 4.4) alloys of variable stochiome-
try that provided intensity and potential values consistent with
those reported in literature [31]. The sharpness of the peaks sug-
gests the formation of large clusters of Sn and the presence of
two-phase regions that caused the capacity to fade [32]. The
decrease in peak intensity on cycling was less marked in the
Mo-containing samples (Fig. 7b), which provides evidence of
the increased reversibility of the reaction.
Ex situ SEM images were obtained upon the addition or
removal of lithium from the Sn (sample A) and Mo0.26Sn0.74
(sample D) electrodes (Fig. 8). A change in the pristine
morphology of the particles was observed when both electrodes
were discharged up to 0.0 V (Fig. 8a and b). LixSn alloys grow
as large mismatched aggregates the size of which is difficult
to estimate. Round Mo particles were not so clearly observed

A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
3397
Table 2
ac impedance parameters (ꢀ) for samples A and D
Sample
Re
R1
Q1 × 10−3
n1
R2
Q2 × 10−3
n2
Q2 × 10−3
n3
A
D
8.7
9.3
59.2
32.1
5.51
8.87
0.67
0.66
52.6
37.2
7.69
7.91
0.65
0.66
4.24
4.04
0.69
0.73
ified the crystal growth rate of Sn and resulted in a markedly
decreased particle size. Also, it increased the reactivity of the
mixtures towards oxygen as revealed by the XPS concentration
depth profile analysis. The effects of Mo on the performance
of the Li/LiPF6 EC-DEC/MoxSn1−x cells can be summarized
as follows, (i) avoidance of the irreversible electrolyte decom-
position observed as an extended plateau at 1.4–1.6 V in pure
Sn; (ii) increased lithium uptake, y, by Sn via the reaction
Sn + yLi ⇔ LiySn, which occurs over the 1.0–0.0 V range and
hence an increased discharge capacity of the electrode; (iii)
increased reversibility of this reaction and hence better capac-
ity retention of the cell. SEM images and impedance spectra
for the electrodes shed some light on the origin of the effects
of Mo. Thus, this element reduces cell impedance as a result
of its increasing the electrical conductivity by filling the cracks
between isolated tin aggregates and hinders SEI formation at
the beginning of the electrochemical process. This must facil-
itate diffusion of lithium species in the Sn particles and result
in improved cell performance. Thus, the use of Mo particles
as an inert additive for Sn electrodes is an attractive choice
to enhance their electrochemical behavior as anodes in lithium
cells. However, further work is required in order to raise the
specific capacity delivered by MoxSn1−x and other mixtures
of metals electrochemically active towards lithium, particularly
as regards preparation of well-dispersed systems consisting of
nanometric particles.
Fig. 9. (a) Impedance spectra for samples A (Sn) and D (Mo0.26Sn0.74) in the
charged state at 1.0 V after the fifth cycle. (b) Equivalent circuit used to fit them.
thebulkparticles. Alternativemodelsassignthehigh-to-medium
frequency semicircle to charge-transfer and the mid-frequency
semicircle to the incorporation of lithium into the lattice [35].
The equivalent circuit used to fit the impedance spectra is shown
in Fig. 9b, where R represents a resistance and Q a constant
phase element. The impedance parameters obtained are listed
in Table 2. The electrolyte resistance (Re) was similar for both
electrodes, but the presence of Mo caused a significant decrease
in R1 and R2. These results illustrate the two beneficial effects
of Mo, namely: improved interparticle conductivity and reduced
electrolyte decomposition and a decreased thickness of the SEI
filmasaresult. Also, theymayaccountfortheimprovedcapacity
retention of the electrodes made from Mo-containing samples
on cycling.
Acknowledgements
´
This work was funded by Junta de Andalucıa (Group FQM-
175) and Spain’s Ministry of Science and Technology (Project
MAT2002-04477-C02-02).
References
[1] J. Wang, I.D. Raistrick, R.A. Huggins, J. Electrochem. Soc. 133 (1986)
457.
[2] I.A. Courtney, J.R. Dahn, J. Electrochem. Soc. 144 (1997) 2943.
[3] L.Y. Beaulieu, K.W. Eberman, R.L. Turner, L.J. Krause, J.R. Dahn,
Electrochem. Solid-State Lett. 4 (2001) A137.
[4] R.A. Huggins, in: J.O. Besenbard (Ed.), Handbook of Battery Materials,
Wiley–VCH, Weinheim, 1999.
4. Conclusions
[5] A.H. Whitehead, J.M. Elliot, J.R. Owen, J. Power Sources 81–82 (1999)
33.
[6] N. Li, C. Martin, J. Electrochem. Soc. 148 (2001) A164.
[7] J. Yang, M. Wachtler, M. Winter, J.O. Besenhard, Electrochem. Solid
State Lett. 2 (1999) 161.
[8] O. Mao, J.R. Dahn, J. Electrochem. Soc. 146 (1999) 423.
[9] X.W. Zhang, C. Wang, A.J. Appleby, F.E. Little, J. Power Sources 109
(2002) 136.
[10] Y. Wang, J.Y. Lee, T.C. Deivaraj, J. Electrochem. Soc. 151 (2004)
A1804.
The effect of Mo particles on the reaction of lithium in Sn
electrodes and the electrochemical behavior were evaluated.
A simple chemical reduction procedure was used to obtain
MoxSn1−x mixtures (0 < x < 0.26) by reducing tin(IV) chloride
in the presence of Mo powder. Metallic Sn and Mo phases were
only detected in the XRD patterns for the MoxSn1−x samples.
EDAX mapping data revealed a uniform distribution of both
elements in the composite material. The presence of Mo mod-

3398
A. Ferna´ndez et al. / Electrochimica Acta 51 (2006) 3391–3398
[11] I. Kim, G.E. Blomgren, P.N. Kumta, Electrochem. Solid State Lett. 7
(2004) A44.
[23] H. Willeemen, T.S. Vandervondel, G.T. Van der Kelen, Inorg. Chim.
Acta. 34 (1977) 175.
[12] G.X. Wang, J. Yao, J. Ahn, H.K. Liu, S.X. Dou, J. Appl. Electrochem.
34 (2004) 187.
[24] J.A. Rodriguez, T. Jirsak, S. Chaturvedi, J. Hrbek, J. Am. Chem. Soc.
120 (1998) 11149.
[13] Y. Liu, J.Y. Xie, Y. Takeda, J. Yang, J. Appl. Electrochem. 32 (2002)
687.
[25] S.D. Beattie, T.D. Hatchard, A. Bonakdarpour, K.C. Hewitt, J.R. Dahn,
J. Electrochem. Soc. 150 (2003) A701.
[14] K.D. Kepler, J.T. Vaughey, M. Thackeray, Electrochem. Solid State Lett.
2 (1999) 307.
[26] L.Y. Beaulieu, S.D. Beattie, T.D. Hatchard, J.R. Dahn, J. Electrochem.
Soc. 150 (2003) A419.
[15] O. Mao, J.R. Dahn, J. Electrochem. Soc. 146 (1999) 423.
[16] H. Kim, Y.J. Kim, D.G. Kim, H.J. Sohn, T. Kang, Solid State Ionics
144 (2001) 41.
[17] Y.L. Kim, H.Y. Lee, S.W. Jang, S.J. Lee, H.K. Baik, Y.S. Yoon, Y.S.
Park, S.M. Lee, Solid State Ionics 160 (2003) 235.
[18] H. Mukaibo, T. Sumi, T. Yokoshima, T. Momma, T. Osaka, Electrochem.
Solid State Lett. 6 (2003) A218.
[27] N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitami, M. Kamino, I. Yonezu,
J. Power Sources 107 (2002) 48.
[28] J. Wang, P. King, R.A. Huggins, Solid State Ionics 20 (1985)
185.
[29] D. Aurbach, A. Nimberger, B. Markovsky, E. Levi, E. Sominski, A.
Gedaken, Chem. Mater. 14 (2002) 4155.
[30] A. Bonakdarpour, K.C. Hewitt, R.L. Turner, J.R. Dahn, J. Electrochem.
Soc. 151 (2004) A470.
[31] I.A. Courtney, J.R. Dahn, J. Electrochem. Soc. 144 (1997) 2045.
[32] I.A. Courtney, J.R. Dahn, J. Electrochem. Soc. 144 (1997) 2943.
[33] H. Li, X. Huang, L. Chen, J. Power Sources 81–82 (1999) 340.
[34] C. Wang, A.J. Appleby, F.E. Little, J. Power Sources 93 (2001)
174.
[19] N. Tamura, M. Fujimoto, M. Kamino, S. Fujitani, Electrochim. Acta 49
(2004) 1949.
[20] J. Morales, L. Sa´nchez, J. Electrochem. Soc. 146 (1999) 1640.
[21] M. Martos, J. Morales, L. Sa´nchez, J. Mater. Chem. 12 (2002) 2979.
[22] H.P. Klug, L.E. Alexander, X-Ray Diffraction Procedures for Polycrys-
talline and Amorphous Materials, John Wiley & Sons, New York, 1974,
p. 665.
[35] H. Li, Z. Lu, X. Huang, L. Huang, L. Chen, Ionics 2 (1996) 259.