Dynamic Article LinksC<Journal ofMaterials Chemistry

Cite this: J. Mater. Chem., 2011, 21, 10042

www.rsc.org/materials PAPER

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online / Journal Homepage / Table of Contents for this issue

Hollow a-LiVOPO4 sphere cathodes for high energy Li-ion batteryapplication

Kuppan Saravanan,a Hwang Sheng Lee,b Mirjana Kuezma,b Jagadese J Vittal*a and Palani Balaya*b

Received 17th December 2010, Accepted 18th February 2011

DOI: 10.1039/c0jm04428h

Hollow spheres of electroactive a-LiVOPO4 were synthesized via a simple one step solvothermal

method. A powder X-ray diffraction study revealed that the obtained product crystallized in the

triclinic a-LiVOPO4 phase. The morphology of the product was largely influenced by reaction

conditions such as reaction time, temperature, etc., and the product morphology was easily fine tuned

from hollow spheres to hard spheres upon changing the reaction time. Without any post-heat treatment

or milling with conductive additives, these hollow spheres exhibited comparatively large reversible Li

storage of 130 and 61 mA h g�1 at 0.1 and 1.7 C respectively. Excellent capacity retention and long term

cycling stability were demonstrated by the hollow spheres of a-LiVOPO4. We believe that a-LiVOPO4

is likely to be a prospective cathode material for high-voltage Li ion batteries application.

Introduction

Ever growing population, industrialisation, concurrent CO2

emission associated with global warming and limited availability

of fossil-fuel together have forced the way for the evolution of

green/alternative technologies of energy production, such as

wind, solar, tidal and wave. However, superior energy storage

systems are mandatory to back up these intermittent energy

sources for a reliable energy supply. Among the existing storage

systems, lithium ion batteries (LIBs) are the most promising

storage device, which has dominated the portable/mobile elec-

tronics industry until now. LIB technology is one of the serious

contenders for powering the automotive electric vehicles and

hybrid electric vehicles.1,2 Ever since LIBs were introduced in the

marketplace, more attention has been paid to focus on improving

the energy/power density. Further efforts have been made to

replace the current layered toxic heavy metal oxide cathodes with

safer, cheaper and less environmental impact electrode mate-

rials.3 Phosphate-based polyanionic transition metal complexes

LiMPO4,3–5 Li3M2(PO4)3,6–12 LiVOPO4F13,14 and LiVOPO415–18

are the best alternatives for the transition metal oxides. Phospho-

olivines possess a robust three dimensional framework, due to

the PO43� polyanion, as strong P–O covalent bonds hinder the

liberation of oxygen. These features provide exceptional stability

for the battery under abnormal conditions. Even though these

phospho-olivines are cheaper, environmentally benign and

exhibit good energy density, they suffer from poor electronic and

aNational University of Singapore (NUS), 3 Science Drive 3, Singapore117543. E-mail: chmjjv@nus.edu.sgbDepartment of Mechanical Engineering, NUS, Singapore 119260. E-mail:mpepb@nus.edu.sg; Fax: +65-6775-4710

10042 | J. Mater. Chem., 2011, 21, 10042–10050

ionic conductivity.19–23 Extensive work has been carried out to

improve the above cited major setbacks.24

In the hunt for perfect polyanion intercalation hosts, lithium

vanadyl phosphate (LiVOPO4) is also considered as a potential

contender. It has a theoretical capacity of �166 mA h g�1 which

is quite closer to LiFePO4 (170 mA h g�1) in addition to this, it

shows higher lithium intercalation potential of nearly 4 V. The

high theoretical energy density (166 mA h g�1 � 3.9 V ¼ 647 Wh

kg�1) with appropriately high operating voltage makes it to be

a fascinating alternate for high volt cathode material in LIBs.

Based on the spatial arrangement of VO6 octahedra and PO4

tetrahedra units which host the Li ions in the interstitial sites of

the framework, LiVOPO4 exists mainly in two different crystal-

lographic phases, namely orthorhombic b-LiVOPO4 and triclinic

a-LiVOPO4.17 Between these phases, b-LiVOPO4 is studied

extensively in the view of lithium ion intercalation, and it shows

better storage performance in comparison with the triclinic a-

LiVOPO4 phase.25 However, single crystal studies by Lii et al.26

show that the structure of a-LiVOPO4 is a close-packed column

of VO octahedra with all interstitial holes alternately filled with

Li and P atoms. Whereas in the b-LiVOPO4 the interstitial hole is

filled with a P atom and a Li atom is located in between the VO

and PO4 coordination polyhedrons surrounded by six oxygen

atoms leading to a distorted octahedron. As a result, b-LiVOPO4

structure does not accomplish the steric condition required

for facile Li ion transport. Besides this, the high temperature

a-LiVOPO4 phase has more open framework with large Li sites in

which Li atoms are loosely bound as compared to the metastable

b-LiVOPO4.26 Till now LiVOPO4 has been synthesised by high

temperature ceramic routes,16,27–29 hydrothermal methods18,26

or combination of these methods to synthesize VOPO4 fol-

lowed by lithiation to produce LiVOPO4.16,17 However, these

methods suffer from the unavoidable high-energy utilization

This journal is ª The Royal Society of Chemistry 2011

Fig. 1 PXRD patterns of the triclinic a-LiVOPO4, s unidentified

impurity.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

and polydispersed growth of the grains due to the high processing

temperatures (generally at 600–900 �C) which also increase the

material cost. Most of the LiVOPO4 synthesised by these methods

resulted in the particle size of around 2–6 or above 6 mm and they

exhibit the poor material utilization during cycling.15,30 From an

energy economy point of view, there is a growing concern to

synthesize the electrode materials via eco-friendly processes, such

as low temperature solvothermal processes.31,32 Here we present

a simple solvothermal approach to synthesize the high tempera-

ture phase of LiVOPO4, namely, triclinic a-LiVOPO4. The judi-

cious choice of experimental parameters helps to control the

morphology of triclinic LiVOPO4 from hollow to hard spheres.

These hollow a-LiVOPO4 spheres display superior electro-

chemical properties compared to the micron sized particles

reported earlier.

Experimental section

All the solvents and chemicals are commercially available and

used as received unless otherwise stated.

Synthesis of a-LiVOPO4

a-LiVOPO4 was synthesized in a single step by the solvothermal

method. A mixture of vanadium(III) acetylacetonate (V(acac)3,

Aldrich), lithium hydroxide (LiOH, Aldrich), ammonium dihy-

drogen phosphate (NH4H2PO4, Aldrich) and ascorbic acid

(C6H8O6, Aldrich) in a 1 : 3 : 1.5 : 2 millimolar ratio was placed

in a Teflon-lined stainless steel reaction vessel. Ethylene glycol

(25 mL) was added as solvent and the vessel was sealed tightly.

The mixture was autoclaved at 300 �C for 10 h in an oven and

then it was allowed to cool naturally to ambient temperature.

The green precipitate obtained was washed copiously with

ethanol and then dried. The synthesized a-LiVOPO4 powders

were directly used to evaluate their electrochemical performance,

without any post-heat treatment.

Physicochemical and electrochemical characterization

Powder X-Ray diffraction (PXRD) patterns were recorded using

a D5005 Bruker X-ray diffractometer equipped with Cu Ka

radiation. The accelerating voltage and current were 40 kV and

40 mA, respectively. A scan speed of 0.015� s�1 was used to

record the PXRD patterns. The morphology of the product was

examined using a field emission scanning electron microscope

(FESEM) model Jeol JSM-6700F operated at 5 kV and 10 mA

and a high resolution transmission electron microscope (JEOL

JEM-2010). For SEM examination, the sample surface was

sputtered with platinum. For TEM studies, the sample was

dispersed in ethanol by sonication, a drop was loaded on a Cu-

grid and dried. For electrochemical studies, composite electrodes

were fabricated with the active material, super P carbon black

and binder (Kynar 2801) in the weight ratio 70 : 15 : 15 using

N-methyl pyrrolidone (NMP) as solvent. Electrodes with

a thickness of 10 mm were prepared using an etched aluminium

foil (20 mm thick) as a current collector using the doctor-blade

technique. Lithium metal foil, 1 M LiPF6 in ethylene carbonate

(EC), diethyl carbonate (DEC) and dimethyl carbonate (DMC)

(1 : 1 : 1, v/v) (Merck) and Celgard 2502 membrane were used as

counter electrode, electrolyte and separator respectively to

This journal is ª The Royal Society of Chemistry 2011

assemble coin-type cells (size 2016) in an Ar-filled glove box

(MBraun, Germany). The geometrical area of the electrode was

2.0 cm2. The active material content in the electrode was around

�1.5 to 2 mg. Further details of cell fabrication have been

described previously.33 The cells were aged for 12 h before

measurement. Charge–discharge cycling at a constant current

mode was carried out using a computer controlled Arbin battery

tester (Model, BT2000, USA) and cyclic voltammetry studies

were carried out at room temperature using a computer

controlled VMP3 (Bio-logic, France). Raman spectra were

recorded with a Raman spectrometer JYT64000.

Results and discussion

PXRD was used to characterize the crystalline phase formation of

a-LiVOPO4 synthesized by the solvothermal method. It is clear

from the PXRD pattern (Fig. 1) that the triclinic a-LiVOPO4

phase was formed. All the peaks in the PXRD pattern were indexed

to a triclinic phase. The structural refinement of the PXRD pattern

was done based on a triclinic structure using P�1 space group.

Obtained lattice parameters for the a-LiVOPO4 spheres are

a ¼ 6.9089 A, b ¼ 7.3009 A and c ¼ 7.8492 A much closer to the

previous report (JCPDS Card No: 72-2253). It can be seen that

very minor unidentified impurities denoted by the symbol s were

present at an angle of 42.9� and 52.2�. No other common impu-

rities such as Li3PO4, Li2VPO6 and V2O5 have been observed.18,29

Several control try-outs have been conducted to establish the

factors that govern the formation of hollow a-LiVOPO4 micro-

spheres including effects of (1) temperature, (2) time, (3)

precursor and (4) solvent.

(1) Effect of temperature

The pure phase was observed only when the temperature was

above 300 �C. Li3PO4 was formed when the temperature is below

300 �C. Fig. 2 shows the PXRD pattern of the Li3PO4 powder

formed during synthesis at 260 �C and 280 �C.

(2) Effect of time

Fig. 3 presents the FESEM images of a-LiVOPO4 synthesised at

various time durations. Pure phase microspheres were observed

J. Mater. Chem., 2011, 21, 10042–10050 | 10043

Fig. 2 PXRD patterns of the products at various temperatures and time.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

when the reaction duration has reached at 20 h (Fig. 3a–c). The

low magnification FESEM image shows the sphere-like archi-

tectures of a-LiVOPO4 ranging from 1–1.5 mm in diameters

(Fig. 3a). The higher magnification FESEM image reveals that

a-LiVOPO4 microspheres are built from small 2D nanoplates

with thickness ranging from 80–120 nm (Fig. 3c). These nano-

plates are aligned interpenetrative to the spherical surface,

directed towards the core of the sphere. Most of the spheres are

broken at the center (Fig. 3b and c). When the reaction

temperature is increased to 30 h microspheres started to elongate

and became irregular in shape and size. (Fig. 3d–f) With further

time progression to 40 h nanoplates fused together to form the

rice ball shaped more dense spheres (Fig. 3g–i).

(3) Effect of precursor

While H3PO4 was used as a phosphate source, nanoplates were

formed. However, they were densely aggregated in the core and

irregular in shape and size, which were also confirmed from the

TEM analysis (see Fig. 4).

(4) Effect of solvent

The solvent has a vital role to play on the morphology of

a-LiVOPO4. We found that only micron sized diamond shaped

particles (Fig. 5a–c) were formed, instead of hollow spheres,

when tetra ethylene glycol was used as solvent instead of ethylene

glycol (EG). This clearly shows the significance of EG in the

formation of LVOP nanoplates and their hierarchical assembly.

Special physical and chemical properties of ethylene glycol, such

as viscosity, vapor pressure, and chelation resulted in the LVOP

crystal growth. Besides, the hydrogen bonding in EG molecules

helps them to exist in long chains which traps the cations (Li and

V4+) in the reaction mixture and may assist in nucleation and

10044 | J. Mater. Chem., 2011, 21, 10042–10050

growth of LVOP in plate shape. The unique chelating ability of

the EG makes it not only a solvent but also a soft template for

assembling the nanoplates into microstructures.34

The TEM, HRTEM and SAED (selected area electron

diffraction pattern) images of hollow a-LiVOPO4 spheres are

shown in Fig. 6. TEM images (Fig. 6a–d) show the hollow nature

of the a-LiVOPO4 synthesised at 20 h. The HRTEM image

(Fig. 6e) of hollow a-LiVOPO4 spheres exhibits clear lattice

fringes demonstrating single crystallinity of the individual plates.

The observed width, 3.271 A, of neighbouring lattice fringes

corresponds to the (�121)/(1–12) plane of a-LiVOPO4. Various

electron diffraction spots shown in Fig. 6f were indexed to

(�121)/(1�12), (402), (2�12) and (�140) plane of a-LiVOPO4. These

SAED spots are consistent with the PXRD pattern shown in

Fig. 1. This confirms that these hollow a-LiVOPO4 spheres are

indeed a pure crystalline phase.

(5) Carbon coating

Diffuse reflectance spectra analysis and DFT calculation by

Yang et al.30 demonstrated that triclinic a-LiVOPO4 is a wide

band gap semiconductor (2.78 eV) which is confirmed using the

AC impedance technique.30 Highly aggregated and dense nature

of the large a-LiVOPO4 particles synthesised commonly by

ceramic methods15,30 exhibit extremely poor electronic conduc-

tivity. Ball milling with acetylene black and adding electronic

binder found to boost the electrochemical and kinetics

process.35,36 In the case of b-LiVOPO4, RuO2 has also been

employed to enhance the electrical conductivity problems,

however, which is expensive for practical applications.25

We have shown that optimising the carbonising precursor will

have strong influence on the storage behaviour of LiFePO4.31,37

Here, we have used ascorbic acid as a carbonising agent. Fig. 7

shows a characteristic Raman spectrum of a-LiVOPO4 hollow

spheres synthesised by the solvothermal method. A small band

appears at 939 cm�1 is attributed to the symmetric stretching

mode of PO43� anion in a-LiVOPO4. Two broad and strong

bands situated at 1357 and 1601 cm�1 are designated to the D

(disordered) and G (graphene) bands of the residual carbon

coated on the hollow spheres respectively. The relative intensity

ratio between the D and G bands can be used to assess the

content of sp3 and sp2 carbon in the a-LiVOPO4. In general,

decreased D/G ratio or increased amounts of sp2 type carbon

greatly enhances the electronic conductivity leading to the good

discharge capacities and superior rate capabilities.38,39 The ID/IG

ratios of the LFP nanoplates were found to be 0.660, this shows

the larger amount of graphene clusters than the disordered

carbon structure, which in turn expected facilitates a far better

cell performance of a-LiVOPO4 when compared with the bulk

sample prepared from ceramic routes.30 Currently systematic

work is in progress on optimising the carbon coating for

a-LiVOPO4.

Electrochemical characterisation

The electrodes made of hollow spheres of a-LiVOPO4 were

subjected to galvanostatic charge–discharge cycles between 3 and

4.5 V vs. Li/Li+ at various current rates. Few charge–discharge

cycles at 0.1 C are shown in Fig. 8. Cycle life of the hollow sphere

This journal is ª The Royal Society of Chemistry 2011

Fig. 3 FESEM images of a-LiVOPO4 synthesized at 300 �C in different time intervals: (a–c) 20 h, (d–e) 30 h and (g–i) 40 h.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

electrode at 0.1 C is shown as inset in Fig. 8. Lithium ion inter-

calation process in the a-LiVOPO4 is expressed in eqn (1).

Reversible lithium extraction and insertion of a-LiVOPO4 are

based on the V4+/V5+ redox couple.17

LiVOPO4 # VOPO4 + Li+ + e� (1)

During the first Li-deintercalation (oxidation) process, the

voltage increased sharply to 4.06 V from the open circuit voltage

(OCV z 3.0 V) followed by a flat long plateau (Fig. 8) and then

gradually increased to the cut-off voltage (4.5 V). The first charge

process resulting in a storage capacity of 145 mA h g�1 which is

equivalent to 87.3% of the theoretical capacity. (Theoretical

capacity 166 mA h g�1 assuming complete extraction of Li.) The

discharge curve shows a similar bi-phasic plateau region at 4.00

V and leading to the storage capacity of 129.5 mA h g�1. In the

first cycle, there is an irreversible capacity loss of about 15 mA h

This journal is ª The Royal Society of Chemistry 2011

g�1 with coulombic efficiency of 90%. Subsequent cycles show the

similar plateau behaviour, however, there is a decline of revers-

ible capacity values for the initial cycles resulting in the invariable

capacity of 101 mA h g�1 at 50th cycle. The small polarisation

(DV) of 0.06 V (DV ¼ voltage difference between the charge and

discharge curves) is indicative of the energetic reversibility of the

V4+/V5+ system under the low rate conditions. Under similar rate

conditions of a-LiVOPO4 prepared by Ren et al.18 exhibited

a huge irreversible capacity loss (ICL) with a coulombic effi-

ciency of 52% and a polarisation of nearly 200 mV. While

comparing with the previously reported triclinic a-LiVOPO4

samples obtained by the sol–gel method,30 hollow a-LiVOPO4

spheres synthesized here exhibit excellent cyclability and very

good rate performance.

To further explain the influence of current rate on the redox

behavior, several replicate cells were tested at different C rates.

The galvanostatic charge–discharge profiles of a-LiVOPO4

hollow spheres at various rates are displayed in Fig. 9a–e.

J. Mater. Chem., 2011, 21, 10042–10050 | 10045

Fig. 4 FESEM and TEM images of a-LiVOPO4 synthesized using H3PO4.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

Polarisation between the charge–discharge plateaus increased

upon increasing the rate. It is found to be 0.08, 0.15, 0.22, 0.35

and 0.45 V for 0.36 C (60 mA g�1), 0.72 C (120 mA g�1), 1.08 C

(180 mA g�1), 2.53 C (420 mA g�1) and 4.22 C (700 mA g�1)

respectively. Columbic efficiency is greatly enhanced to 98%

beyond 0.36 C. The capacity retention of the a-LiVOPO4 elec-

trodes versus cycling number is shown in Fig. 9f which reveals

good and consistent cycling stabilities found at all the rates. The

overall reversible storage capacity is found to be 85, 77, 71, 48

and 40 mA h g�1 for 0.36 C (60 mA g�1), 0.72 C (120 mA g�1),

1.08 C (180 mA g�1), 2.53 C (420 mA g�1) and 4.22 C (700 mA

g�1) respectively. As compared to the chemically lithiated and

mechanically ground a-LiVOPO4 by Kerr et al.17 at 1 C rate,

Fig. 5 FESEM images of the a-LiVOPO4 synthesised in te

10046 | J. Mater. Chem., 2011, 21, 10042–10050

these a-LiVOPO4 hollow spheres display a better performance. It

is believed that the hollow nature favours efficient wetting of the

active materials by electrolytes thereby providing more active

sites for the electrochemical reactions.40 Thus the morphology of

the product appears to be one of the important factors in

improving the kinetic properties of the a-LiVOPO4.

Galvanostatic cycling of the a-LiVOPO4 sample at 1.7 C rate is

shown in Fig. 10. Under this rate (280 mA g�1) 500 cycles were

carried out to elucidate the long term cycling stability. Reversible

capacities of 58 mA h g�1 and 51 mA h g�1 were observed at the

end of 1st and 500th cycle respectively. This shows an average

capacity degradation of just 0.01 mA h g�1 per cycle. In addition

to this, the observed columbic efficiency was found to be more

traethylene glycol as solvent at different magnifications.

This journal is ª The Royal Society of Chemistry 2011

Fig. 6 TEM, HRTEM and SAED images of hollow spheres of a-LiVOPO4.

Fig. 7 Raman spectrum of the a-LiVOPO4 hollow spheres.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

than 99% during these long term cycles. The long term stability of

the lithium insertion/extraction reactions in a-LiVOPO4 hollow

spheres is evidenced by negligible capacity fade over the 500

charge–discharge cycles. Plate type architecture offers the

advantages of fast transport of Li+ ions. Besides this, the hollow

nature of the spheres offers sufficient active sites and short

diffusion path for lithium ions to intercalate. These features lead

to good electrochemical activity of hollow a-LiVOPO4 spheres

compared with the bulk (>6 mm) LiVOPO4 synthesised from

ceramic routes.18,30,35

This journal is ª The Royal Society of Chemistry 2011

The high rate capability is one of the mandatory electro-

chemical features of LIBs to power the high energy applications

(HEV and EV). The rate performance of a-LiVOPO4 electrodes

was evaluated for 9 different (Fig. 11) charge–discharge current

rates corresponding to 0.18, 0.6, 1.7, 3.4, 5.1, 6, 8.5, 13 and 0.36 C

(here 1 C ¼ 166 mA g�1), in the voltage range 3–4.5 V (Fig. 11).

On average, minimum 50 cycles were carried out for a given rate.

As expected, the capacity decreases from 98 mA h g�1 to 22 mA h

g�1 with increasing C-rate from a value of 0.18 C (16 mA g�1) to

13 C (2158 mA g�1), showing a diffusion-limited mass transfer of

Li+ between the surface and core of the a-LiVOPO4 particles.

Upon decreasing the current from 13 C to 0.36 C, 85 mA h g�1

was observed. At each C-rate, the storage capacity is found to be

stable except at a low rate of 0.18 C. Fig. 11 shows charge–

discharge voltage profiles of hollow a-LiVOPO4 spheres at

various C rates. Upon increasing the C rate, it is seen that the

polarization of the electrode material increases, which imitates

the variation of DV, this is possibly due to increase in the elec-

trode resistance, thereby declining the storage performance of the

electrodes.

The cyclic voltammograms (CV) of hollow a-LiVOPO4

spheres are shown in Fig. 12. The CV was recorded with Li metal

as the counter and reference electrodes in the voltage range of

3–4.5 V at the scan rate of 0.058 mV s�1 up to 10 cycles at room

temperature. A single pair of anodic/cathodic peaks found in the

CV which corresponds to the V4+/V5+ redox couple trans-

formations. During the first cycle (i.e., Li-deintercalation), the

anodic peak is at 4.06 V and the corresponding cathodic peak is

J. Mater. Chem., 2011, 21, 10042–10050 | 10047

Fig. 8 Galvanostatic charge–discharge cycle curves of a-LiVOPO4 hollow spheres at 0.1 C. Current density of 16 mA g�1, 1 C refers to a capacity of 166

mA g�1 in one h, potential window 3–4.5 V and the data were recorded at room temperature (inset: capacity vs. cycle number plots).

Fig. 9 Galvanostatic charge–discharge profiles of a-LiVOPO4 hollow spheres at various rates (selected cycles are given): (a) 0.36 C (60 mA g�1); (b) 0.72 C

(120 mA g�1); (c) 1.08 C (180 mA g�1); (d) 2.53 C (420 mA g�1); (e) 4.22 C (700 mA g�1) and (f) reversible capacity vs. cycle number plots, here open symbols refer

to the charge capacity and closed symbols refer to the discharge capacity. The potential window is 3–4.5 V and the data were recorded at room temperature.

10048 | J. Mater. Chem., 2011, 21, 10042–10050 This journal is ª The Royal Society of Chemistry 2011

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

Fig. 10 (a) Long term galvanostatic charge–discharge voltage profiles of

a-LiVOPO4 at 1.7 C. (b) Reversible capacity vs. cycle number plots

(selected cycles are given) over the potential window 3–4.5 V.

Fig. 12 Cyclic voltammograms of a-LiVOPO4 hollow spheres. Scan

rate: 0.058 mV s�1.

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

at 3.85 V (i.e., Li-intercalation) in good agreement with the

charge/discharge curves (Fig. 8.). The symmetrical nature of

the redox peaks in the CV is suggestive of the good movement of

the interfacial boundary.17 In the subsequent cycles, the oxida-

tion peak and the corresponding reduction peak are unaltered

upto 10 cycles; this infers good reversibility of a-LiVOPO4. This

also suggests that the unidentified impurities observed in the

PXRD (Fig. 1) were electrochemically inactive in the potential

window used. In comparison, the results of solvothermally

Fig. 11 a-LiVOPO4 hollow spheres: (a) the rate capability and (b)

charge–discharge voltage profiles at various C rates.

This journal is ª The Royal Society of Chemistry 2011

prepared a-LiVOPO4 hollow spheres indicate that this synthetic

approach offers significant improvement in producing an elec-

trode material with favorable electrochemical properties over the

existing methods.18,30,35

Conclusions

In summary, triclinic a-LiVOPO4 hollow spheres were syn-

thesised by the solvothermal method at 300 �C for 10 h and they

were characterized using PXRD, SEM, HRTEM and SAED.

Reaction parameters have a strong influence on the morphology

of the final product. The as-synthesized a-LiVOPO4 was tested as

a 4 V cathode material for LIB and it exhibited reversible

capacities of 130 and 61 mA h g�1 at 0.1 C and 1.7 C respectively.

The remarkable long term cycling stability and good high rate

performance were found upto 13 C. The hollow nature and

nanosized plates forming the spheres favour the absolute wetting

of the a-LiVOPO4 by the liquid electrolyte so that Li+ interca-

lation/deintercalation can be achieved easily leading to the

superior Li storage when compared with the a-LiVOPO4 syn-

thesised by other methods. Therefore, hollow a-LiVOPO4

spheres synthesised by the solvothermal method is a very good

alternative for the cathode materials in certain LIB applications.

Acknowledgements

We thank the Ministry of Education Singapore for financial

support through NUS FRC Grant No. R-143-000-371-112 and

US DARPA grant R-265-000-320-597. K.S. thanks NUSNNI

for research scholarship. H.S.L thanks Andrew A. O. Tay for

fruitful discussions.

References

1 M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657.2 J. Hassoun, S. Panero, P. Reale and B. Scrosati, Adv. Mater., 2009,

21, 4807–4810.3 A. K. Padhi, K. S. Nanjundaswamy and J. B. Goodenough, J.

Electrochem. Soc., 1997, 144, 1188–1194.4 S. Y. Chung, J. T. Bloking and Y. M. Chiang, Nat. Mater., 2002, 1,

123–128.5 H. Huang, S. C. Yin and L. F. Nazar, Electrochem. Solid-State Lett.,

2001, 4, A170–A172.

J. Mater. Chem., 2011, 21, 10042–10050 | 10049

Publ

ishe

d on

24

Mar

ch 2

011.

Dow

nloa

ded

by T

he U

nive

rsity

of

Auc

klan

d L

ibra

ry o

n 29

/10/

2014

00:

11:0

7.

View Article Online

6 A. S. Andersson, B. Kalska, P. Jonsson, L. Haggstrom, P. Nordblad,R. Tellgren and J. O. Thomas, J. Mater. Chem., 2000, 10, 2542–2547.

7 Y. Z. Li, Z. Zhou, M. M. Ren, X. P. Gao and J. Yan, Electrochim.Acta, 2006, 51, 6498–6502.

8 C. Masquelier, C. Wurm, J. Rodriguez-Carvajal, J. Gaubicher andL. Nazar, Chem. Mater., 2000, 12, 525–532.

9 S. Patoux, C. Wurm, M. Morcrette, G. Rousse and C. Masquelier, J.Power Sources, 2003, 119, 278–284.

10 M. M. Ren, Z. Zhou, Y. Z. Li, X. P. Gao and J. Yan, J. PowerSources, 2006, 162, 1357–1362.

11 H. Huang, S. C. Yin, T. Kerr, N. Taylor and L. F. Nazar, Adv.Mater., 2002, 14, 1525–1528.

12 M. Y. Saidi, J. Barker, H. Huang, J. L. Swoyer and G. Adamson,Electrochem. Solid-State Lett., 2002, 5, A149–A151.

13 J. Barker, R. K. B. Gover, P. Burns and A. Bryan, Electrochem. Solid-State Lett., 2005, 8, A285–A287.

14 J. Barker, R. K. B. Gover, P. Burns, A. Bryan, M. Y. Saidi andJ. L. Swoyer, J. Power Sources, 2005, 146, 516–520.

15 B. M. Azmi, T. Ishihara, H. Nishiguchi and Y. Takita,Electrochemistry, 2003, 71, 1108–1110.

16 J. Barker, M. Y. Saidi and J. L. Swoyer, J. Electrochem. Soc., 2004,151, A796–A800.

17 T. A. Kerr, J. Gaubicher and L. F. Nazar, Electrochem. Solid-StateLett., 2000, 3, 460–462.

18 M. M. Ren, Z. Zhou, X. P. Gao, L. Liu and W. X. Peng, J. Phys.Chem. C, 2008, 112, 13043–13046.

19 C. Delacourt, L. Laffont, R. Bouchet, C. Wurm, J. B. Leriche,M. Morcrette, J. M. Tarascon and C. Masquelier, J. Electrochem.Soc., 2005, 152, A913–A921.

20 P. P. Prosini, M. Lisi, D. Zane and M. Pasquali, Solid State Ionics,2002, 148, 45–51.

21 R. Amin, J. Maier, P. Balaya, D. P. Chen and C. T. Lin, Solid StateIonics, 2008, 179, 1683–1687.

22 M. S. Islam, D. J. Driscoll, C. A. J. Fisher and P. R. Slater, Chem.Mater., 2005, 17, 5085–5092.

10050 | J. Mater. Chem., 2011, 21, 10042–10050

23 C. A. J. Fisher and M. S. Islam, J. Mater. Chem., 2008, 18, 1209–1215.24 A. Yamada, S. C. Chung and K. Hinokuma, J. Electrochem. Soc.,

2001, 148, A224–A229.25 M. M. Ren, Z. Zhou, L. W. Su and X. P. Gao, J. Power Sources, 2009,

189, 786–789.26 K. H. Lii, C. H. Li, C. Y. Cheng and S. L. Wang, J. Solid State Chem.,

1991, 95, 352–359.27 H. T. Kuo, N. C. Bagkar, R. S. Liu, C. H. Shen, D. S. Shy,

X. K. Xing, J. F. Lee and J. M. Chen, J. Phys. Chem. B, 2008, 112,11250–11257.

28 M. M. Ren, Z. Zhou and X. P. Gao, J. Appl. Electrochem., 2010, 40,209–213.

29 K. Nagamine, T. Honma and T. Komatsu, J. Am. Ceram. Soc., 2008,91, 3920–3925.

30 Y. Yang, H. S. Fang, J. Zheng, L. P. Li, G. S. Li and G. F. Yan, SolidState Sci., 2008, 10, 1292–1298.

31 K. Saravanan, P. Balaya, M. V. Reddy, B. V. R. Chowdari andJ. J. Vittal, Energy Environ. Sci., 2010, 3, 457–464.

32 K. Saravanan, M. V. Reddy, P. Balaya, H. Gong, B. V. R. Chowdariand J. J. Vittal, J. Mater. Chem., 2009, 19, 605–610.

33 K. Saravanan, K. Ananthanarayanan and P. Balaya, Energy Environ.Sci., 2010, 3, 939–948.

34 Y. Wang, X. Jiang and Y. Xia, J. Am. Chem. Soc., 2003, 125, 16176–16177.

35 B. M. Azmi, T. Ishihara, H. Nishiguchi and Y. Takita, J. PowerSources, 2005, 146, 525–528.

36 B. M. Azmi, H. S. Munirah, T. Ishihara and Y. Takita, Ionics, 2005,11, 402–405.

37 K. Saravanan, J. J. Vittal, M. V. Reddy, B. V. R. Chowdari andP. Balaya, J. Solid State Electrochem., 2010, 14, 1755–1760.

38 M. M. Doeff, J. D. Wilcox, R. Kostecki and G. Lau, J. PowerSources, 2006, 163, 180–184.

39 M. Doeff, J. Wilcox, R. Yu, A. Aumentado, M. Marcinek andR. Kostecki, J. Solid State Electrochem., 2008, 12, 995–1001.

40 M. Nagarathinam, K. Saravanan, W. L. Leong, P. Balaya andJ. J. Vittal, Cryst. Growth Des., 2009, 9, 4461–4470.

This journal is ª The Royal Society of Chemistry 2011