Structural and electrochemical properties of Al3+ doped V 2O5 nanoparticles prepared by an oxalic acid assisted soft-chemical method

Article abstract of DOI:10.1016/j.jallcom.2010.03.133

V₂O₅ and Al_0.2V₂O₅ nanoparticles were prepared by an oxalic acid assisted soft-chemical method. X-ray photoelectron spectroscopy confirmed the V⁵⁺ oxidation state of V₂O₅, whereas an intermediate state between V ⁵⁺ and V⁴⁺ of Al_0.2V₂O₅. Raman scattering showed that the Al³⁺ ions existed in an [AlO ₆] octahedral environment. The doping of Al³⁺ increased the cohesion between the V₂O₅ slabs, which enhanced the structural stability of the material. The chemical diffusion coefficients of the Al_0.2V₂O₅ nanoparticles were a little bit smaller than those of V₂O₅. Charge-discharge cycling showed that the Al_0.2V₂O₅ nanoparticles exhibited much better capacity retention than the un-doped V₂O ₅, which was attributed to the enhanced structural stability of the material.

Full text of DOI:10.1016/j.jallcom.2010.03.133

Journal of Alloys and Compounds 502 (2010) 92–96
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
3
+
Structural and electrochemical properties of Al doped V O nanoparticles
2
5
prepared by an oxalic acid assisted soft-chemical method
a
b,∗
a
b
b
b
a
Shiying Zhan , Yingjin Wei , Xiaofei Bie , Chunzhong Wang , Fei Du , Gang Chen , Fang Hu
a
College of Materials Science and Engineering, Jilin University, Changchun 130012, China
College of Physics and State Key Laboratory of Superhard Materials, Jilin University, Changchun 130012, China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
2 5 2 5
V O and Al0.2V O nanoparticles were prepared by an oxalic acid assisted soft-chemical method. X-ray
Received 16 August 2009
Received in revised form 15 March 2010
Accepted 16 March 2010
Available online 20 March 2010
5+
photoelectron spectroscopy confirmed the V oxidation state of V2O5, whereas an intermediate state
5+
4+
3+
between V and V of Al0.2V2O5. Raman scattering showed that the Al ions existed in an [AlO6] octahe-
3+
dral environment. The doping of Al increased the cohesion between the V2O5 slabs, which enhanced the
structural stability of the material. The chemical diffusion coefficients of the Al0.2V2O5 nanoparticles were
a little bit smaller than those of V2O5. Charge–discharge cycling showed that the Al0.2V2O5 nanoparticles
exhibited much better capacity retention than the un-doped V2O5, which was attributed to the enhanced
structural stability of the material.
Keywords:
Nanostructured materials
V2O5
Raman scattering
Electrochemistry
Rechargeable lithium batteries
©
2010 Elsevier B.V. All rights reserved.
of Fe3+ and Cr3+ doped V O5 using X-ray diffraction [11,14]. It is
1
. Introduction
Vanadium oxides (e.g. VO , V O and V O5) are potential cath-
2
revealed that the [MO ] octahedral in the doped materials link the
6
2
6
13
2
V O slabs, which increase the three-dimensional character of the
2
5
ode materials for rechargeable lithium batteries [1–3]. Among
material. This enhances the stability of the crystal structure and
improves the electrochemical performance of the material. Raman
scattering, which is much sensitive to the short-range environment
of coordinative units, has been widely used in structural analysis.
However, up to now there are limited works to study the local struc-
ture of cation doped V O using Raman scattering. In this work,
these materials, V O5 is probably the most studied one due to its
2
unique features such as easy preparation, low cost, high stabil-
ity and large energy density [4]. However, it is well known that
pristine bulk V O5 is not an appropriate cathode material because
2
of its low lithium diffusion coefficient and poor structural stabil-
ity with lithium insertion/extraction, which lead to bad battery
performance such as poor capacity retention and low rate capa-
bility. Recently, many studies have been focused on the composite
2
5
we prepared Al3+ doped V O5 nanoparticles using an oxalic acid
2
3+
assisted soft-chemical method. We studied the effects of Al dop-
ing on the structural properties of V O5 using Raman scattering.
2
or novel structured V O5 cathode materials [5–8]. Especially, the
In addition, the electrochemical properties of the material were
2
nanostructured materials exhibit better electrochemical perfor-
compared with those of un-doped V O5 nanoparticles with the
2
mance with respect to bulk V O5 by virtue of their morphology
aim to find a promising cathode material for rechargeable lithium
batteries.
2
properties. The small size of these materials can boost redox reac-
tivity and their short diffusion pathways enable the materials to
withstand high discharge rates over a long cycle life.
2
. Experimental
Besides these, some other studies showed that the electrochem-
3+
For the preparation of Al doped V2O5 nanoparticles, 0.01 mol of V2O5 (99%,
ical performance of V O5 could be improved by introducing some
Junsei) was dissolved in 100 ml of 0.3 mol L−1 oxalic acid solution. Afterwards, a
0.1 mol L Al(NO3)3 was slowly titrated into the above solution until a molar ratio
of Al:V = 1:10 was reached. The solution was stirred for 5–6 h at 80 C, followed
by drying at 100 C to get a homogeneous precursor. This precursor was then sub-
jected to heat treatment at 350 C to obtain the final product. The preparation of
V2O5 nanoparticles followed the same procedure as above but without using of the
Al(NO3)3 solution.
2
+
2+
3+
3+
−1
guest cations (M) such as Ag , Cu , Fe and Cr into the material
matrix [9–13]. The improvement in electrochemical performance
◦
◦
of the cation doped V O5 has been attributed to their stabled crystal
2
◦
structure. Baffier and co-workers studied the structural properties
Elemental analysis of Al and V was performed on an ICP-AES/1000 (PE Company)
inductively coupled plasma instrument. The morphology features of the material
were studied by scanning electron microscope (JEOL JSM-6700F). X-ray photoelec-
tron spectroscopy (XPS) was performed on an ESCALAB spectrometer (VG scientific)
∗ Corresponding author. Tel.: +86 431 85155126; fax: +86 431 85155126.
E-mail address: yjwei@jlu.edu.cn (Y. Wei).
0
925-8388/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2010.03.133

S. Zhan et al. / Journal of Alloys and Compounds 502 (2010) 92–96
93
using a monochromic Mg K␣ excitation. The binding energy scale was corrected
using the C 1s peak at 284.8 eV. X-ray diffraction data were collected on a Bruker D8
diffratometer with Cu K␣ radiation. Raman scattering was performed on a Renishaw
RM1000 micro-Raman instrument. The excitation light source was an Ar-ion laser
with ꢀ = 514.5 nm.
Electrochemical experiments were carried out using a two-electrode battery
cell, using metallic lithium foil as the anodic electrode. The working electrode was
composed of 75 wt.% of active material, 15 wt.% of carbon black and 10 wt.% of
poly-vinylidenefluoride (PVDF) dissolved in N-methylpyrrolidone (NMP). The slurry
mixture was spread on an Al foil and then dried in vacuum oven. The fabricated
2
cathode electrode was cut into a size of 0.64 cm (0.8 × 0.8). The electrolyte was
a 1 M lithium perchlorate (LiClO4) in 1:1 (v/v) ethylene carbonate/diethyl carbon-
ate (EC/DEC). Galvanostatic charge–discharge cycling was carried out on a Land®
(
Wuhan) automatic battery cycler in the potential window of 4.0–2.0 V. Cyclic
®
voltammetry was collected on a ZAHNER -IM6e electrochemical workstation in
the potential window of 4.0–2.0 V.
3
. Results and discussion
3.1. Structural and morphology analysis
During the preparation of V O5 nanoparticles, the V O5 raw
2
2
material was dissolved in oxalic acid solution. In the mean while,
the solution color changed to blue. This occurred because oxalic
acid is a reductive agent, which caused the reducing of V5 to V
+
4+
according to the following reaction,
V O5 + 3H C O → 2[VO(C O )](blue) + CO + 3H O
2
2
2
4
2
4
2
2
4
+
The V ions have a typical color of blue. Subsequently, small
clusters containing V4 ions were formed during the reaction of
+
V O5 with oxalic acid. When the precursor was heat treated in air,
2
4+
5+
the V ions were oxidized to V with the simultaneous decom-
position of the precursor. This led to the formation of fine V O5
2
nanoparticles which was confirmed by SEM as shown in Fig. 1(a).
In addition, the SEM image of the Al3 doped V O5 (Fig. 1(b)) shows
+
2
Fig. 1. SEM images of the (a) V2O5 and (b) Al0.2V2O5 nanoparticles.
that the material was also composed of nano-sized particles, but
with significant particle agglomeration. Elemental analysis deter-
mined that the molar ratio of Al:V of the material was close to
−1
at 992, 691 and 527 cm were assigned to the stretching modes
of the V–O(1), V–O(3) and V–O(2) bonds, respectively. The bands
1
:10.
Fig. 2 shows the V 2p3/2 XPS spectra of the as-prepared mate-
rials. The V 2p3/2 spectrum of the V O5 sample was characterized
−1
located at 404 and 280 cm corresponded to the bending modes of
2
V–O(1), and those observed at 478 and 297 cm were assigned to
the bending modes of V–O(3) and V–O(2), respectively. There were
two bands recorded at 191 and 140 cm , which corresponded to
the [VO5]–[VO5] vibrations. These vibrations called as “external
−1
5+
by a single peak centered at 517.6 eV, which corresponds to the V
oxidation state [15]. The Al3 doped sample showed a main peak at
+
−1
5
17.6 eV, with a small shoulder at 516.3 eV. The later value fit well
4
+
with that of V in VO2 [16]. This indicates that the vanadium ions
in the material were in an intermediate oxidation state between
V
modes” reflected the cohesion between the V O5 slabs.
2
5+
4+
4+
and V . The existence of V ions was usually observed in
cation doped V O5 materials such as Fe V O5.16, Cu0.04V O5 and
2
0.12
2
2
AgxV O5 gels [9,11,17]. Since it is difficult to determine the exact
2
oxidation state of V in the material, we simplified the chemical
composition of the Al3+ doped V O5 as Al V O5 in the following
2
0.2
2
text.
Fig.
3
shows the X-ray diffraction patterns of the as-
prepared nanoparticles. There were no apparent differences
between the two diffraction patterns. This indicates that the
crystal structure of Al0.2V O5 was close to that of orthorhom-
2
bic V O5, which was consistent with the previous report
2
by Baffier et al. [18]. The lattice parameters of the materi-
als were calculated based on the Pmmn space group, which
were a = 11.507(4) Å, b = 3.565(1) Å, c = 4.378(2) Å for V O5 and
2
a = 11.511(3) Å, b = 3.555(1) Å, c = 4.359(3) Å for Al0.2V O5. The val-
2
ues in brackets are estimated standard deviations. The results
indicate that the Al3+ dopant in V O5 caused slight lattice expansion
2
along the a axis and small compression along the b and c axis.
Fig. 4 displays the Raman patterns of the materials. There are
three different V–O bonds in the V O5 structure, i.e. the terminal
2
V–O(1) bond, the chaining V–O(2) bond and the bridging V–O(3)
Fig. 2. V 2p2/3 XPS spectra of the V2O5 and Al0.2V2O5 nanoparticles.
bond [19]. As for the V O5 nanoparticles, the Raman bands recorded
2

9
4
S. Zhan et al. / Journal of Alloys and Compounds 502 (2010) 92–96
Fig. 3. X-ray diffraction patterns of the V2O5 and Al0.2V2O5 nanoparticles.
It is very difficult to identify the Al–O vibrations from the Raman
pattern of Al0.2V O5. However, we do observe a significant increase
2
−
1
in the Raman intensity between 450 and 600 cm . This could be
due to the Al–O vibrations of the [AlO ] octahedral [20], which were
6
likely overlapped with the V–O vibrations. Most of the V–O bands of
Al0.2V O5 located at the same wavenumbers as those of V O5. How-
2
2
−
1
ever, the [VO5]–[VO5] vibration increased from 140 to 145 cm
after Al3+ doping, which indicates that the cohesion between the
3+
V O5 slabs was strengthened by Al doping. This also means that
2
3+
the structural stability of the material was enhanced by Al dop-
ing. This would be good for the electrochemical performance of the
material from the structural point of view. In addition, the V–O(3)
stretching mode shifted from 691 to 701 cm , corresponding to
the softening of the V–O(3) bond. There were no obvious changes
on the V–O(1) and V–(O2) vibrations. This indicates that the Al3
cations did not significantly affect the V–O(1) and V–(O2) bonds.
−1
+
Fig. 5. Cyclic voltamgrams of the (a) V2O5 and (b) Al0.2V2O5 nanoparticles.
+
3.2. Electrochemical properties
reversible structural transitions during Li insertion/extraction. The
CV behavior of the V O5 and Al V O5 nanoparticles was different
2
0.2
2
Fig. 5 shows the cyclic voltamgrams (CV) of the materials at dif-
from that of crystalline bulk V2O5, which is known to experi-
ence irreversible structural transitions in the potential window of
4.0–2.0 V [21]. Recent studies have shown that these irreversible
structural transitions can be effectively overcome by Cu and Cr
doping [10,12]. The chemical diffusion coefficients of the materi-
−1
ferent scan rates from 50 to 200 ␮V s . Both materials exhibited
ꢀ
ꢀ
three well-resolved redox couples, which are labeled as a/a , b/b
ꢀ
2+
3+
and c/c , respectively. This indicates that the materials underwent
als, D , can be evaluated from the CV measurements. In the case of
Li
semi-infinite diffusion, the peak current Ip can be expressed by the
Randles and Sevcik equation [22,23]:
Ip = 2.687 × 10 n3/2ꢁ1/2
5
D
1/2
Li
ACLi
(1)
where n is the number of electrons per species reaction, ꢁ the
scan rate, A the electrode area, CLi the Li+ concentration in the
electrode. Fig. 6 plots the Ip dependence on the scan rate of the
CV measurement, from which Ip exhibited a linear relationship
with the square root of scan rate (ꢁ1 ). Using this relationship
/2
and the Eq. (1), the DLi values of the V O5 nanoparticles at the
2
−
10
−11
peak-a, -b and -c were calculated to be 7.0 × 10
, 3.8 × 10
−
12
2
−1
and 2.1 × 10
of the Al V O5 nanoparticles were 2.6 × 10
cm s , respectively. The corresponding values
−
10
−11
and
, 2.4 × 10
cm s , respectively, which were a little smaller than
0
.2
2
−
13
2
−1
9
.6 × 10
3+
those of V O5. This may be due to the Al dopants, which block the
2
+
Li diffusion in the material lattice.
Galvanostatic charge–discharge cycling was carried out in
the potential window of 4.0–2.0 V. Fig. 7 displays the typical
charge–discharge profiles of the materials at the current density
Fig. 4. Raman spectra of the V2O5 and Al0.2V2O5 nanoparticles.

S. Zhan et al. / Journal of Alloys and Compounds 502 (2010) 92–96
95
Fig. 8. Cycling performance of the V2O5 and Al0.2V2O5 nanoparticles at the current
−
1
density of 150 mA g
.
Fig. 6. The Ip vs. ꢁ1/2 relationships of the (a) V2O5 and (b) Al0.2V2O5 nanoparticles.
of 150 mA g−1. Both materials showed three discharge plateaus
at ∼3.3, 3.1 and 2.3 V, which were consistent with the cyclic
0.2 2 5
Fig. 9. Cycling performance of the Al V O nanoparticles at different current den-
−
1
voltamgrams. However, the discharge plateaus of the Al0.2V O5
sities from 100 to 400 mA g
.
2
nanoparticles were much shorter than those of V O5. In addi-
2
tion, a gradual potential decrease was observed between 3.1 and
be done to investigate the phase transitions of the materials dur-
ing lithium intercalation. Fig. 8 shows the cycling performance of
2
.3 V, in contrast to the abrupt decrease of the V O5 nanoparti-
2
cles. This suggests limited structural changes taken place in the
the materials. The initial discharge capacity of the V O5 nanoparti-
2
Al0.2V O5 nanoparticles. Precise structural determinations should
−1
2
cles was ∼260 mAh g . The material showed continuous capacity
−
1
fading, which dropped below 140 mAh g
after 50 cycles. The
Al0.2V O5 nanoparticles exhibited a smaller discharge capacity of
2
−
1
∼
220 mAh g in the first cycle. This was probably due to the addi-
tional Al cations, which may occupy partial of the crystallographic
sites that originally belong to Li . The material showed a capacity
3+
+
loss in the second cycle. However, it is exciting to see that the mate-
rial showed rather good cycling stability with prolonged cycling.
−
1
The discharge capacity was 180 mAh g
after 50 cycles, which
exhibited almost no capacity fading with respect to the third cycle.
In addition, the material exhibited good rate capability in that a
−
1
reversible discharge capacity as high as 150 mAh g was obtained
−
1
when the current density increased to 400 mA g (Fig. 9).
4. Conclusions
We have prepared V O5 and Al V O5 nanoparticles by a sim-
2
0.2
2
ple soft-chemical method. The Al0.2V O5 nanoparticles had an
2
3+
orthorhombic structure as that of V O5. The Al ions were bonded
2
with oxygen in an [AlO ] octahedral environment. The doping
of Al increased the cohesion between the V2O5 slabs, which
Fig. 7. Charge–discharge potential profiles of the V2O5 and Al0.2V2O5 nanoparticles
6
−1
3+
at the current density of 150 mA g
.

9
6
S. Zhan et al. / Journal of Alloys and Compounds 502 (2010) 92–96
enhanced the structural stability of the material. Even though the
[5] J. Gao, J. Kim, A. Manthiram, Electrochem. Commun. 11 (2009) 84–86.
[
6] W. Wu, Y. Wang, X. Wang, Q.Q. Chen, X. Wang, S.Y. Yang, X.M. Liu, J. Guo, Z.H.
Yang, J. Alloys Compd. 486 (2009) 93–96.
chemical diffusion coefficients of Al0.2V O5 were a little bit smaller
2
than those of the V O5 nanoparticles, the doped material still
2
[7] S.L. Chou, J.Z. Wang, J.Z. Sun, D. Wexler, M. Forsyth, H.-K. Liu, D.R. MacFarlane,
S.-X. Dou, Chem. Mater. 20 (2008) 7044–7051.
showed better electrochemical performance, especially excellent
capacity retention. The reason for the improved electrochemical
[
8] C.Q. Feng, S.Y. Wang, R. Zeng, Z.P. Guo, K. Konstantinov, H.K. Liu, J. Power Sources
84 (2008) 485–488.
9] F. Coustier, S. Passerini, W.H. Smyrl, Solid State Ionics 100 (1997) 247–258.
1
3
+
performance by Al doping was at least attributed to the enhance-
ment in structural stability of the material.
[
[10] M. Giorgetti, M. Berrettoni, W.H. Smyrl, Chem. Mater. 19 (2007) 5991–6000.
[
11] J. Farcy, S. Maingot, P. Soudan, J.P. Pereira-Ramos, N. Baffier, Solid State Ionics
9 (1997) 61–69.
12] C. Leger, S. Bach, P. Soudan, J.-P. Pereira-Ramos, Solid State Ionics 176 (2005)
365–1369.
[13] S.Y. Zhan, G. Chen, D.L. Liu, A. Li, C.Z. Wang, Y.J. Wei, J. Alloys Compd. 479 (2009)
52–656.
9
Acknowledgements
[
1
This work was supported by MOST (No. 2009CB220104)
and NSFC (No. 50702024). The authors thank the Program for
Changjiang Scholar and Innovative Research Team in University
6
[
14] G. Grégoire, N. Baffier, A. Kahn-Harari, J.-C. Badot, J. Mater. Chem. 8 (1998)
2103–2108.
(
(
No. IRT0625) and the Program for New Century Excellent Talents
NCET-07-0366) for financial supporting.
[15] G. Bliznakov, Y. Pesheva, D. Klissurski, M. Marinov, V. Kozhukharov, Appl. Catal.
9 (1987) 211–218.
16] T. Christmann, B. Felde, W. Niessner, D. Schalch, A. Schrmann, Thin Solid Films
87 (1996) 134–138.
[17] Y.J. Wei, C.W. Ryu, K.B. Kim, J. Power Sources 165 (2007) 386–392.
2
[
2
References
[
[
18] N. Baffier, L. Znaidi, M. Huber, Mater. Res. Bull. 25 (1990) 705–713.
19] R. Baddour-Hadjean, J.P. Pereira-Ramos, C. Navone, M. Smirnov, Chem. Mater.
[
[
1] E. Baudrin, G. Sudant, D. Larcher, B. Dunn, J.-M. Tarascon, Chem. Mater. 18
2006) 4369–4374.
2] H.M. Zeng, Y. Zhao, Y.J. Hao, Q.Y. Lai, J.H. Huang, X.Y. Ji, J. Alloys Compd. 477
2009) 800–804.
20 (2008) 1916–1923.
(
[
20] D.K. Breitinger, H.-H. Belz, L. Hajba, V. Komlósi, J. Mink, G. Brehm, D. Colognesi,
S.F. Parker, R.G. Schwab, J. Mol. Struct. 706 (2004) 95–99.
(
[
[
[
21] F.J. Anaissi, G.J.F. Demets, H.E. Toma, Electrochem. Commun. 1 (1999) 332–335.
22] M.D. Levi, D. Aurbach, J. Electroanal. Chem. 421 (1997) 79–88.
23] F. Wu, L. Wang, C. Wu, Y. Bai, Elctrochim. Acta 54 (2009) 4613–4619.
[
[
3] Y. Chen, H. Liu, W.-L. Ye, Scripta Mater. 59 (2008) 372–375.
4] X. Wei, L. Jiao, S. Liu, J. Sun, W.X. Peng, H. Gao, Y. Si, H. Yuan, J. Alloys Compd.
4
86 (2009) 672–676.