Preparation and characterization of LiNiVO4 powder and non-stoichiometric LiNixVyOz films

Article abstract of DOI:10.1016/j.materresbull.2007.06.023

Inverse spinel type structured oxide, LiNiVO₄, was synthesized by using solid-state method and the crystalline powder was characterized by Rietveld refinement and X-ray photoelectron spectroscopy. Non-stoichiometric lithium nickel vanadate thin films were prepared by physical vapour deposition technique. The amorphous films were characterized by Rutherford back-scattering spectroscopy (RBS), nuclear reaction analysis (NRA), Auger electron spectroscopy (AES), X-ray diffraction (XRD), scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) analytical methods. Films crystal growth at various temperatures was also studied by XRD and SEM. The HRTEM analysis of sputtered film shows nanocrystalline domains of NiO and LiNiVO₄ phases with characteristic lattice parameters of the host compound and the results correlate well with the XRD data. Electrochemical properties of the films were discussed.

Full text of DOI:10.1016/j.materresbull.2007.06.023

Materials Research Bulletin 43 (2008) 1519–1527
www.elsevier.com/locate/matresbu
Preparation and characterization of LiNiVO powder and
4
non-stoichiometric LiNi V O films
x y z
*
M.V. Reddy , C. Wannek, B. Pecquenard, P. Vinatier, A. Levasseur
ICMCB/ENSCPB (CNRS UPR 9048), Universit e´ Bordeaux 1, 87 Avenue du Dr. Schweitzer 33608 Pessac Cedex, France
Received 3 January 2007; accepted 4 June 2007
Available online 12 June 2007
Abstract
Inverse spinel type structured oxide, LiNiVO
4
, was synthesized by using solid-state method and the crystalline powder was
characterized by Rietveld refinement and X-ray photoelectron spectroscopy. Non-stoichiometric lithium nickel vanadate thin films
were prepared by physical vapour deposition technique. The amorphous films were characterized by Rutherford back-scattering
spectroscopy (RBS), nuclear reaction analysis (NRA), Auger electron spectroscopy (AES), X-ray diffraction (XRD), scanning
electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) analytical methods. Films crystal
growth at various temperatures was also studied by XRD and SEM. The HRTEM analysis of sputtered film shows nanocrystalline
domains of NiO and LiNiVO
the XRD data. Electrochemical properties of the films were discussed.
2007 Elsevier Ltd. All rights reserved.
4
phases with characteristic lattice parameters of the host compound and the results correlate well with
#
Keywords: A. Thin films; B. Sputtering; C. Electron diffraction; C. X-ray diffraction; D. Electrochemical properties
1
. Introduction
Recently, metal oxide LiNiVO (bulk and thin film) has been studied extensively in literature as an electrode
4
material [1–10] for Li-ion batteries. Various methods of preparation and characterization are employed on optimizing
morphology, structure and electrochemical performance of this material because it can be used as high-potential
cathode material [1] and also high-capacity anode material [5]. Thin film solid-state batteries are of potential
applications for the use in smart cards, implantable medical devices, CMOS, etc. because of their small ideal
dimensions and high efficiency. In addition to energy storage, thin film materials are used as electrocatalysts,
electrochromics, lubricants and so on.
Chemical, physical, microstructure, electronic and electrical properties of various non-stoichiometric transition
metal oxides, electrode materials preparation and characterization have been studied in literature [10–14]. Recently,
LiNiVO films were prepared by radio frequency magnetron (rf) magnetron sputtering [7–9]. The composition,
4
structure and morphology of films vary with the physical sputtering deposition parameters like power, total pressure
and oxygen partial pressure and distance between target substrate and substrate temperature (in situ or ex situ).
*
Corresponding author. Present address: Department of Physics, National University of Singapore, S12,03, Science Drive 3, Singapore 117542,
Singapore. Tel.: +65 65162605; fax: +65 67776126.
E-mail address: phymvvr@nus.edu.sg (M.V. Reddy).
0025-5408/$ – see front matter # 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.materresbull.2007.06.023

1520
M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
Fundamental and applied studies on non-stoichiometric thin film compositions in the amorphous or nanocrystalline/
polycrystalline state showed interesting structural phases and morphology and physical properties [10–14]. Presently,
we report Rietveld refinement of LiNiVO powder and preparation and characterization of non-stoichiometric films by
4
varying oxygen partial pressure during sputtering.
2
. Experimental
2
.1. Preparation of powder and target
The LiNiVO crystalline powder was prepared by solid-state method. A stoichiometric mixture of high-purity
4
(
Aldrich) Li CO , NiO (prepared from decomposition of Ni(NO ) 6H O) and V O is heated at 550 8C for 2 h to
2
3
3 2
2
2 5
eliminate carbonates, finally at 730 8C in air for 12 h at a heating rate of 3 8C/min. Final single phase product was used
in fabrication of target. The target with a 5 mm diameter and 3 mm thickness was used for rf sputtering and target was
prepared by using ꢀ20 g of LiNiVO powder and 5% stearin binder mixed well and finally cold pressed at ꢀ260 MPa
4
pressure. In order to remove binder and improve the robustness, the target was sintered at 600 8C for 8 h in a box
furnace, in air.
Fig. 1. (a) Rietveld-refined XRD pattern (continuous lines, top) of LiNiVO
4
powder, vertical bars represent positions of allowed (h k l) lines and
structure. Li, Ni ions are in octahedral site and V ions in
difference pattern is also shown. (b, c) Crystallographic representation of LiNiVO
tetrahedral site.
4

M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
1521
2
.2. Thin films preparation and characterization
The radio frequency (rf) sputtering technique (physical vapour deposition technique) was used to fabricate lithium
nickel vanadate films. The sputtering unit was placed in a dry-Argon glove box, to avoid contamination. The pure
Argon (99.99%) was used as carrier gas, before sputtering of films; vacuum of ꢀ4 ꢁ 10ꢂ5 Pa was applied into
sputtering chamber for overnight. The sputtering conditions are: power, 30 W; oxygen partial pressure (PO ), 0 and
100 (10%) mPa gas; distance between target and substrate, 8 cm; total working pressure of Ar, 1 Pa; substrates are not
heated during sputtering. Presputtering was carried out under identical conditions for 2 h duration to remove
2
contaminated top target surface. For clarity thin films prepared at P : 0 mPa refer to compd. I and P : 100
O
2
O
2
(
10%) mPa refer to compd. II. Thin films were deposited on various substrates depending on characterization method.
Carbon (RBS/NRA) and stainless steel substrates (for electrochemical studies) were polished to get mirror finish by
using silicon carbide paper. Finally, substrates were cleaned in distilled water and acetone by using ultrasonic bath,
dried in an oven and transferred into a glove box. Silicon (1 1 1) and aluminium foil were used for SEM studies. High-
˚
resolution TEM measurements were carried out with the thin films (thickness ꢀ200 A) deposited directly on copper
grid during sputtering. Here, we show TEM studies of only amorphous films prepared at PO ¼ 0 mPa and annealed
2
films were not studied due to oxidation of Cu grids above 300 8C and samples are very fragile. In case of the annealed
films, XRD studies give useful information about the structure of the films. The crystalline films are obtained by
sintering the amorphous non-stoichiometric films deposited on Si substrates at various temperatures from 200 to
9
00 8C, for 2 h duration in a tubular furnace.
x y z
Fig. 2. RBS spectrum of LiNi V O films of: (a) compd. I and (b) compd. II. Symbols: Open squares represent experimental RBS data and
continuous lines represent simulated data by RUMP. Elements and substrate peaks are shown in figures.

1522
M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
4
+
The compositions of thin films were studied by Rutherford back-scattering technique (RBS), by using 2 MeV, He
ion beam. The experimental RBS spectrum was simulated by the use of Rutherford Universal Manipulation Program
7
RUMP) software. The lithium content was analysed by nuclear reaction analysis (NRA) ( Li(pa) He). The RBS and
4
(
NRA measurements were performed at Nuclear Research Centre (CENBG), University of Bordeaux. The Auger
electron spectroscopy (AES) (VG Micro lab 310F) was used to check depth profile of the films and also to give
homogeneity. The structures of powder and thin films were studied by using X-ray diffractometer (XRD, Philips
PW1730) with Cu Ka radiation. Rietveld refinement of powder was performed by using Fullprof software. X-ray
photoelectron spectroscopy (XPS) analysis of the powder was performed by Surface Science Instruments
spectrometer (model 301). Scanning electron microscopy (SEM) (JEOL JSM-5200 microscope) was used to study the
surface morphology of the films. High-resolution electron microscopy (HRTEM) images were taken by using JEOL
3
010 microscope. The electrochemical discharge–charge cycling was carried by using a galvanostic mode, 1 M LiPF6
was used as electrolyte and cells were cycled in the range of 0.02–3.0 V, Li metal was used as counter and reference
electrode.
3
. Results and discussion
3
.1. Characterization of LiNiVO powder
4
The obtained LiNiVO powder was in yellow colour. The Rietveld refinement X-ray diffraction pattern of the
4
¯
powder is shown in Fig. 1. The experimental spectra were refined with the space group Fd3m (2 2 7). The crystal
2
parameters of the LiNiVO powder are as follows: Li at the 16d site (5/8, 5/8, 5/8), occupancy 0.5, Beq (A ), 1.97(3);
˚
4
Fig. 3. AES atomic concentration vs. depth profile of: (a) compd. I and (b) compd. II.

M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
1523
2
Ni at the 16d site (5/8, 5/8, 5/8), occupancy 0.5, Beq (A ), 2.02(5); V at the 8a site (0,0,0), occupancy 1.0, Beq (A ),
2
˚
˚
2
.78(3); O at the 32e site (0.3700(2), x, x), occupancy 0.5, Beq (A ), 2.24(6). The obtained lattice parameter value of
˚
1
P
P
1=2
2
˚
LiNiVO powder was a = 8.21844(5) A, wRp ¼ 0:1118, R = 0.0340, where wRp ¼ ½ wiðy ꢂ y Þ2= wjyio
ꢃ
4
B
io
ic
P
P
and R = jI ꢂ I j/ I , where y and y represent the number of counts observed and calculated, wi the
B
Ko
Kc
Ko
io
ic
statistical factor, I and I are the observed and calculated intensity. The crystallographic representation of LiNiVO
o
c
4
˚
structure is shown in Fig. 1b and c. The lattice parameter of the compound agrees well with Bernier et al. (a = 8.215 A)
˚
15] and Liu et al. (a = 8.223 A) [4]. The X-ray photoelectron spectroscopy studies of LiNiVO showed binding
4
[
energy of Ni 2P3/2, 855.1 eV; V 2P3/2, 516 eV; Li 1s, 54.6 eV; O 2P, 529.6 eV. The SEM photograph shows (Fig. 1a)
agglomerated particles with sizes ranging from 6 to 10 mm.
3
.2. Chemical composition of thin films
Rutherford back-scattering spectrometry has been widely used for surface analysis by nuclear physicists [16]. It
provides a rapid and simple method of examining films purity, composition and thickness. The experimental RBS
spectra are shown in Fig. 2a and b. For convenience, compd. I (P : 0 mPa) and compd. II (P : 100 mPa) refer to
O
O
2
2
former and later composition, respectively. The RBS spectra clearly show a peak corresponding to energy channel
Fig. 4. X-ray diffraction pattern of: (a) as-deposited film and (b) annealed films of compd. I and (c) compd. II annealed film with Cu Ka radiation.
Film thickness t > 1 mm; silicon (1 1 1) substrate; 2 h annealing time and temperature are indicated in figures.

1524
M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
numbers of Ni, V, O and C elements, where C peak is due to the substrate and absence of other surface impurities in the
spectra. The simulated spectra (continuous line) by RUMP software of the compds. I and II are shown in Fig. 2a and b.
The chemical composition of compds. I and II were LiNi V
O
and LiNi V
1
O
1
0.75 (ꢄ0.03) 3.4(ꢄ0.1)
1.6 (ꢄ0.03) 5.8(ꢄ0.1)
respectively and these values are in good agreement with peaks area (Ni, V, O) (Fig. 2). The lithium content was not
determined using RBS due to its low surface scattering and matrix scattering height, these values of LiNiVO film
4
were reported elsewhere [6]. The NRA analysis shows that lithium contents of compds. I and II were Li1
,
and Li1.3
respectively. AES plots of atomic concentration versus depth profiles of compds. I and II are shown in Fig. 3. The
(ꢄ0.1)
(ꢄ0.1)
sputtered films are homogeneous with thickness, this shows uniform deposition of elements (Ni, V, O) during
sputtering time irrespective of deposition conditions, except for compositional (atomic concentration) variation. The
Fig. 5. SEM photographs of films of: (a) compd. I, amorphous, bar scale, 1 mm; (b) compd. I, annealed films at 730 8C, bar scale 2.0 mm; (c) compd.
II, annealed at 730 8C, bar scale 4.0 mm; (d) same as (c) in expanded scale; (e) compd. II, annealed at 900 8C, bar scale, 4.0 mm. Annealing time, 2 h.

M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
1525
lithium depth profile of the non-stoichiometric films with thickness by NRA analysis is not studied, we assume from
our previous studies on Li concentration analysis of stoichiometric films that Li-concentrations are in good
homogeneity with the thickness of the film, more details on analysis are discussed previously [6].
3
.3. Crystal structure of films
The colour of deposited films varies depending on film thickness, it varies from green, brown, yellow and black
>1 mm). The XRD patterns of the as-deposited sputtered film of compd. I are shown in Fig. 4a. The broad peaks in the
(
XRD pattern are characteristic of amorphous nature of film; broad peaks correspond to NiO (2 0 0) and also LiNiVO4
films (4 0 0). In the amorphous films, it is difficult to differentiate exact phases (h k l) lines [NiO (2 0 0) or LiNiVO4
(
effect of annealing temperature on the growth of the LiNi V O films (compds. I and II) studied by using XRD. The
4 0 0)]. To understand the nature of these (h k l) lines and the structure and crystal growth of the films, we studied the
x
y z
XRD patterns of the compd. I at various annealing temperatures (200–700 8C) are shown in Fig. 4b. Intensity of (2 2 0)
h k l line is much stronger than (1 1 1) line due to the presence of transition metal atom (V) in tetrahedral site, whereas
normal spinel like LiMn O (1 1 1) line is strongest and (2 2 0) is the weakest line. XRD diagrams of the annealed film
2
4
of the compd. I showed a major phase of LiNiVO in addition to NiO, Li VO are present as impurity phases where as
4
4
3
compd. II shows a mixture of LiV O or Li VO and LiNiVO phases are shown in Fig. 4c. The XRD patterns of
2
5
3
4
4
annealed films (compds. I, II) correlate well with the chemical compositions obtained from RBS analysis.
3
.4. Morphology, nanostructure and electrochemical studies
The SEM images of the compds. I and II are shown in Fig. 5. As-deposited films are dense in morphology. The
compds. I and II show as-expected variation of particle size/morphology with annealing temperature. The compd. I
shows submicron-sized particles (Fig. 5b) and interestingly the compd. II shows a rod-like morphology, these films on
0
˚
Fig. 6. (a and a ) High-resolution TEM images of amorphous film of compd. I ꢀ200 A thick sputtered on copper grid; (b) Schematic view of (d1 1 1
)
line in LiNiVO structure.
4

1526
M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
further heating at high temperature, variation of surface morphology, particle size are clearly seen in Fig. 5e. In
addition, the particle size varies with thickness of the films, smaller thickness (ꢅ0.3 mm) particle is lesser compared to
0
thick films (ꢆ1 mm). The HRTEM images of compd. I are shown in Fig. 6a and a . The photographs clearly show as-
deposited films are in amorphous nature and in addition to two types of nanocrystallite domains are clearly seen in
0
˚
˚
Fig. 6a . The nanodomain corresponds to lattice spacing of NiO (d) = 2.27 A and LiNiVO (d) = 4.8 (ꢄ0.1) A
4
˚
line of the host LiNiVO (d = 4.75 A) and
compounds. These d-values of compd. I compare well with the d(
1 1 1)
4
schematic diagram (d1 1 1) lines in the host compound are displayed in Fig. 6b. Our studies showed that crystallite
nanodomains vary depending on deposition rate and reactive gas [7,11,12]. In general morphology, crystallite size of
the sputtered films depends on mobility (i.e. low mobility or medium to high mobility) of the atoms [17].
The electrochemical charge–discharge cycles of the compds. I and II are shown in Fig. 7a and b cycled in the
2
potential range of 0.02–3.0Vand at a current density of 10 mA cm . The first discharge cycle plots clearly show voltage
plateaus at ꢀ1.70 V, ꢀ0.75 V and cycling plots are characteristic nature of amorphous films. During first discharge
2
Fig. 7. (a–c) Galvanostatic discharge–charge cycling plots of compds. I and II. Voltage range, 3.0–0.02 V; current density, 10 mA/cm ; geometrical
2
area of the electrode, 1.32 cm . Number in figures represents discharge–charge cycle number.

M.V. Reddy et al. / Materials Research Bulletin 43 (2008) 1519–1527
1527
cycle of the compds. I and II, 4.7–7.5 moles of lithium are inserted into the host compound which corresponds to a
capacity of 690 and 1130 (ꢄ30) mAh/g. The capacity versus cycle number plots of the compds. I and II is shown in
Fig. 7c and compds. I and II deliver a reversible (1st charge) capacity of 550 and 820 mAh/g. Indeed stoichiometric
LiNiVO films deliver a superior capacity depending on structure and morphology and annealing temperature [7,9].
4
The reasons for the variation in electrochemical properties are due to compositional variation, morphology and
structural defects in the films.
4
. Conclusions
In this paper, we prepared LiNiVO powder and characterized it by Rietveld refinement and XPS. LiNi V O thin
4
x y z
films were prepared by using rf magnetron sputtering in absence of oxygen and high partial pressure of oxygen (10%)
during deposition. The compositions of films were systematically studied by RBS, NRA. Sputtered films are
homogeneous with thickness conformed by AES. SEM analysis showed that films morphology varies with sputtering
conditions, annealing temperature and composition of the films. XRD analysis showed as-deposited films are
amorphous and nano/polycrystalline/crystalline films are formed by annealing treatment. HRTEM analysis of the as-
deposited film showed a nanocrystalline nature with a characteristic lattice spacing of the host compound.
Electrochemical properties of films during 1st discharge capacity ranges from 690 to 1130 (ꢄ30) mAh/g.
Acknowledgements
Thanks are due to N. Martz, Department of Material Science, Technical University of Darmstadt, Germany, for help
with TEM; P. Moretto, CENBG, University of Bordeaux, for his help with RBS and NRA measurements; M. Lahaye,
ICMCB-CNRS, for help with AES; D. Gonbeau, LCTPCM, Pau, France, for help with XPS analysis.
References
[
[
[
1] G.T.K. Fey, P. Muralidharan, C.Z. Lu, Y.-D. Cho, Solid State Ionics 177 (2006) 877.
2] A. Subramania, N. Angayarkanni, S.N. Karthick, T. Vasudevan, Mater. Lett. 60 (2006) 3023.
3] Z. Zhongqiang, M. Junfeng, X. Lijin, T. Hua, Z. Jun, H. Yingmo, H. Xiang, W. Pingwei, D. Jinhui, Z. Zhibin, W. Hongfen, C. Haiyan, J. Am.
Chem. Soc. 88 (2005) 2622.
[
[
[
[
[
[
4] R.S. Liu, Y.C. Cheng, R. Gundakaran, L.Y. Jang, Mater. Res. Bull. 36 (2001) 1479.
5] F. Orsini, E. Baudrin, S. Denis, L. Dupont, M. Touboul, D. Guyomard, Y. Piffard, J.-M. Tarascon, Solid State Ionics 107 (1998) 123.
6] C. Rossignol, G. Ouvrard, E. Baudrin, J. Electrochem. Soc. 148 (2001) A869.
7] M.V. Reddy, B. Pecquenard, P. Vinatier, A. Levasseur, J. Phys. Chem. B 110 (2006) 4301.
8] M.V. Reddy, B. Pecquenard, P. Vinatier, C. Wannek, A. Levasseur, P. Moretto, Nucl. Instrum. Methods Phys. Res. B 246 (2006) 397.
9] M.V. Reddy, B. Pecquenard, P. Vinatier, A. Levasseur, J. Power Sources 163 (2007) 1040.
[
[
[
[
[
[
[
[
10] E. Schmidt, F. Weill, G. Meunier, A. Levasseur, Thin Solid Films 245 (1994) 34.
11] I. Martin-Litas, P. Vinatier, A. Levasseur, J.C. Dupin, D. Gonbeau, F. Weill, Thin Solid Films 416 (2002) 1.
12] K.J. Rao, H. Benqlilou-Moudden, G. Couturier, P. Vinatier, A. Levasseur, Mater. Res. Bull. 37 (2002) 1353.
13] I. Martin-Litas, P. Vinatier, A. Levasseur, J.C. Dupin, D. Gonbeau, Bull. Mater. Sci. 26 (2003) 673.
14] K.J. Rao, B. Pecquenard, A. Gies, A. Levasseur, J. Etourneau, Bull. Mater. Sci. 29 (2006) 535.
15] J.C. Bernier, P. Poix, A. Michel, Bull. Soc. Chim. Fr. (1963) 445.
16] K.-P. Lieb, Contemp. Phys. 40 (1999) 385.
17] R.J. Messier, Vac. Sci. Technol. A 4 (1986) 490.