Synthesis and electrochemical properties of high performance yolk-structured LiMn2O4 microspheres for lithium ion batteries

Article abstract of DOI:10.1039/c2ta00204c

Yolk-structured microspheres of spinel LiMn₂O₄ are successfully prepared by a specially designed multi-step synthesis procedure involving precipitation, controlled oxidation, selective etching and chemical lithiation. Solid-structured and hollow-structured LiMn₂O₄ are also synthesized by a similar method for comparison. X-ray diffraction, scanning electron microscopy, transmission electron microscopy, Brunauer-Emmett-Teller method and IR spectroscopy are employed to study their structures and compositions. The electrochemical properties of the LiMn ₂O₄/Li cells are also tested. The results indicate that LiMn₂O₄ powder composed of yolk-structured microspheres possesses remarkable high rate capability and outstanding high capacity retention not only at room temperature but also at elevated temperatures. This study may provide significant new insight into restraining the capacity fading of LiMn₂O₄ electrodes and the yolk-structured LiMn ₂O₄ may be used for the next generation of lithium ion batteries.

Full text of DOI:10.1039/c2ta00204c

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Synthesis and electrochemical properties of high
performance yolk-structured LiMn2O4 microspheres for
Cite this: J. Mater. Chem. A, 2013, 1,
860
lithium ion batteries†
*
Yu Qiao, Si-Rong Li, Yan Yu and Chun-Hua Chen
Yolk-structured microspheres of spinel LiMn2O4 are successfully prepared by a specially designed multi-step
synthesis procedure involving precipitation, controlled oxidation, selective etching and chemical lithiation.
Solid-structured and hollow-structured LiMn2O4 are also synthesized by a similar method for comparison.
X-ray diffraction, scanning electron microscopy, transmission electron microscopy, Brunauer–Emmett–Teller
method and IR spectroscopy are employed to study their structures and compositions. The electrochemical
properties of the LiMn2O4/Li cells are also tested. The results indicate that LiMn2O4 powder composed of
yolk-structured microspheres possesses remarkable high rate capability and outstanding high capacity
retention not only at room temperature but also at elevated temperatures. This study may provide
significant new insight into restraining the capacity fading of LiMn2O4 electrodes and the yolk-
structured LiMn2O4 may be used for the next generation of lithium ion batteries.
Received 6th September 2012
Accepted 25th October 2012
DOI: 10.1039/c2ta00204c
www.rsc.org/MaterialsA
case of large particle sizes. Although the nano-structured
LiMn2O4 is very effective in improving the rate capability, the
1 Introduction
In the past few decades we have witnessed the rapid development
of lithium ion batteries (LIBs) in response to the increasing
needs of energy storage and conversion.1–3 For example, they
have made it possible for traditional gasoline-powered automo-
biles to evolve into a generation of electric vehicles (EV). As the
most suitable substitute to replace the conventional LiCoO2
cathode material, spinel-structured LiMn2O4 has been
spotlighted due to its high power density, natural abundance,
low material cost and environmental benignity.4–7 However, it
exhibits a serious capacity fading during charge and discharge,
especially at elevated temperatures. The fading can be attributed
to several factors such asJahn–Teller distortion due to Mn3+ ions,
spinel dissolution into the electrolyte and electrolyte decompo-
sition in the high potential regions.8–12 Thus, the improvement in
the electrochemical reversibility and stability of LiMn2O4 at high
rates and elevated temperatures becomes more and more
important for scientists.13–15
low tap density of nano-sized powders directly leads to the low
energy density of a cell. To achieve a high tap density, electrode
materials are preferred to have micron-sized particles, espe-
cially with spherical shapes which can pack more densely.21–25 In
our study, we want to nd a balance between high tap density
and short diffusion distance. Inspired by the work of Qian et al.
who have synthesized hollow structures of Mn2O3, MnO2 and
Mn2O3 microspheres,26–28 we design and synthesize a special
yolk-structured LiMn2O4 microsphere (Scheme 1b). In this
structure model, the outer shell is porous and composed of
many nanoparticles, which can enlarge the specic surface area
of the electrode and provide more reaction sites for lithium
insertion and extraction. The core is relatively dense to increase
the volumetric energy density. The spacing between the shell
and the core can buffer any volume change of the core during
heating/cooling or charge/discharge.
In the present work, we rst synthesize a yolk-structured
Mn2O3 as a precursor and then follow a simple solid-state
reaction or chemical lithiation to produce the yolk-structured
LiMn2O4 microsphere (LMO-Y) (Scheme 1a). Moreover, we
compare the electrochemical performance of LMO-Y with those
of hollow LiMn2O4 microspheres (LMO-H) (Scheme 1c) and
solid LiMn2O4 microspheres (LMO-S) (Scheme 1d).
Recently, LiMn2O4 nano-materials with various morphol-
ogies have been prepared to improve the cycle life and rate
capability,16–20 because the poor performance of LiMn2O4 is also
attributed to the long distance for lithium ions to diffuse in the
CAS Key Laboratory of Materials for Energy Conversion, Department of Materials
Science and Engineering, University of Science and Technology of China, Anhui
Hefei 230026, China. E-mail: cchchen@ustc.edu.cn; Fax: + 86 551 3601592; Tel: +
86 551 3606971
2 Experimental
† Electronic supplementary information (ESI) available: Sample surface
morphology, electrode thickness measurement, and three different
yolk-structured LiMn2O4 samples (conditions, structures and cycling properties)
are given. See DOI: 10.1039/c2ta00204c
2.1 Synthesis of MnCO3 microspheres
Manganese sulphate (MnSO4$H2O, 0.507 g) and sodium bicar-
bonate (NaHCO3, 2.32 g) were separately dissolved in 210 ml of
860 | J. Mater. Chem. A, 2013, 1, 860–867
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Scheme 1 Schematic illustration of the synthetic process of a yolk-structured LiMn2O4 microsphere (a) and the sphere structures of yolk-structured (b), hollow (c), and
solid LiMn2O4 (d).
distilled water. About 21 ml of ethanol was then added to the solution was 4.2 mol Lꢁ1 instead of 0.6 mol Lꢁ1 to dissolve
MnSO4 solution under vigorous magnetic stirring and the completely the MnCO3 core and leave only a hollow MnO2 shell
NaHCO3 solution was then directly added to cause precipita- remained aer this step.
tion. The mixture solution was maintained at room temperature
ꢀ
(nearly 25 C) for 3 h. A precipitate of MnCO3 was obtained by
centrifuging and washed with distilled waterꢀ and absolute
alcohol for several times, and then dried at 80 C for 1 h in a
vacuum.
2.4 Synthesis of LMO-S, LMO-Y and LMO-H
The above as-prepared MnCO3 microspheres, yolk-structured
MnCO3 (core)–MnO2 (shell) microspheres and hollow MnO2
microspheres were separately calcined at 530 C for 15 h. The
ꢀ
obtained Mn2O3 powders with different structures were sepa-
rately mixed and ground with lithium acetate (CH3COOLi$H2O)
2.2 Synthesis of yolk-structured MnCO3 (core)–MnO2 (shell)
microspheres
in the stoichiometric molar ratio (Li : Mn ¼ 1 : 2). Then the
ꢀ
The as-prepared MnCO3 (0.2 g) was dispersed in 40 ml of mixtures were calcined at 285 C for 3 h and subsequently cal-
distilled water. Then 20 ml of 0.032 mol Lꢁ1 KMnO4 solution cind at 650 ꢀC for 6 h. All the heat treatment processes were
was added under vigorous stirring for 40 min to form a homo- carried out in air atmosphere. Finally, three spinel LiMn2O4
geneous solution. Then 20 ml of 0.6 mol Lꢁ1 HCl was added to powders with different microstructures, LMO-S, LMO-Y and
the above suspension and the mixture was maintained with LMO-H, were prepared. It should be mentioned that the
stirring for 2 min followed by a rapid centrifuging step. The as- synthesis methods here for LMO-S and LMO-H are similar to
obtained powder was washed with distilled water and absolute previous reports.29,30
alcohol for several times. It was nally dried at 80 ꢀC for 1 h in a
vacuum.
2.5 Morphology and structure characterization
Powder X-ray diffraction (XRD) was performed on a diffrac-
tometer (Philips X’pert Pro Super with Cu Ka radiation) at room
2.3 Synthesis of hollow MnO2 microspheres
The synthetic process was similar to the steps above to prepare temperature. The XRD patterns were recorded in the 2q range
the yolk-structured MnCO3 (core)–MnO2 (shell) microspheres. from 10ꢀ or 20ꢀ to 80ꢀ at a scan rate of 2ꢀ minꢁ1. A scanning
The difference was that the concentration of the added HCl electron microscopy (SEM, JSM-6390 LA, JEOL) study of the
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 860–867 | 861

View Article Online
Journal of Materials Chemistry A
Paper
powders was performed to analyze their morphologies. The LiMn2O4 spheres are prepared by a simple reaction between
samples were analyzed under a Hitachi H-800 transmission Mn2O3 and lithium acetate at a high temperature (650 ꢀC).
electron microscope (TEM) at an accelerating voltage of 200 kV. Again, the LiMn2O4 shell is expected to be porous, which is
Fourier transformation infrared (FTIR) spectroscopy was also conrmed by our experiments (Fig. S1d†).
conducted with a Magna-IR 750 spectrometer in the range of
400–4000 cmꢁ1 with a resolution of 4 cmꢁ1. The specic surface
3.2 Structures of MnCO3, MnCO3 (core)–MnO2 (shell) and
area of the powders was determined by the Brunauer–Emmett–
Teller (BET) method using a Beckmancoulter SA3100 specic
surface area and aperture tester.
LiMn2O4
The crystal structure of the as-synthesized product was exam-
ined by powder X-ray diffraction (XRD) and the result is shown
in Fig. 1. The pattern of the synthesized MnCO3 precursor
(Fig. 1a) can be indexed as a pure phase (JCPDS 44-1472). The
intermediate products MnCO3–MnO2 (Fig. 1b) and MnO2
(Fig. 1c) show only two broad peaks of a poorly crystallized
MnO2, in agreement with the literature.26,28 In Fig. 1d–f, the XRD
patterns of the powders show that all LiMn2O4 samples are well
crystallized and can be identied as a pure spinel LiMn2O4
structure (JCPDS 35-0782).
Fig. 2 shows the particle sizes and morphologies of the
LiMn2O4 samples and their precursors. As shown in these SEM
images, the three LiMn2O4 samples and their corresponding
precursors are all rather homogeneous spherical spheres with a
diameter of about 1.5 mm, which are very useful for further
processing. The TEM images of the LiMn2O4 powders (Fig. 2d–f)
conrm the successful syntheses of their designed microstruc-
tures LMO-S, LMO-Y and LMO-H. In fact, the SEM observations
of some broken spheres in LMO-Y and LMO-H also conrm the
presence of hollow structures. It can also be seen that aer the
calcinations and reaction with lithium acetate, the LiMn2O4
samples (Fig. 2d–f) are found to be smaller than their precursors
(Fig. 2a–c) and the surface becomes rougher, which is probably
caused by the emission of CO2 gas. Particularly, the LiMn2O4
shells in LMO-Y and LMO-H are even rougher than LMO-S,
suggesting that the LiMn2O4 shells are composed of many
nanocrystalline grains. The TEM images also reveal that the
shell thickness of LMO-Y and LMO-H is all nearly 100 nm and
the inner diameter of LMO-Y is about 700 nm.
2.6 Electrochemical performance tests
The LiMn2O4 active materials, acetylene black and poly-
(vinylidene uoride) (80 : 10 : 10, w/w/w) were mixed into slur-
ries in N-methyl-2-pyrrolidone in an agate mortar. They were
then cast on aluminum foils with a doctor blade and dried at 70
ꢀC for 10 h to obtain several electrode laminates. Discs
(diameter ¼ 14 mm) of the laminates were punched, dried at
70 ꢀC for 2 h in a vacuum oven and then were transferred to an
argon-lled glove box (MBraun Labmaster 130). A typical
loading of the active material was 6.0–6.5 mg. Aerwards,
CR2032 type coin-cells with Li as the counter electrode were
assembled in the glove box with the electrolyte of 1 M LiPF6
solution in ethylene carbonate (EC) and dimethyl carbonate
(DMC) (1 : 1, w/w, Zhuhai Smoothway Electronics Materials Co.
Ltd). The cyclic voltammograms (CV) of the cells were measured
on a CHI 604A electrochemical workstation from 3.2 to 4.35 V at
a scan rate of 0.1 mV sꢁ1. The cells were also cycled on a multi-
channel battery test system (Neware BTS-2300, Shenzhen) in the
voltage window from 3.2 to 4.35 V. The room temperature was
controlled at 25 ꢀC (ꢂ3 ꢀC) by an air-conditioner. The high
temperature performances of the cells were also tested by laying
ꢀ
ꢀ
them in an oven at 55 C (ꢂ1 C).
3 Results and discussion
3.1 Design of the synthesis procedure for yolk-structured
LiMn2O4
As shown in Scheme 1a, a simple precipitation step is employed
to prepare dense MnCO3 spheres. Then, a controlled oxidation
by KMnO4 of these MnCO3 spheres is followed to form a MnO2
(shell)–MnCO3 (core) structure. The oxidation reaction can be
written as 2KMnO4 + 3MnCO3 ¼ 5MnO2 + 3CO2 + K2O. In the
solution environment, K2O can react with water to form KOH
which is soluble. Therefore, the products CO2 and K2O can
cause the micro-porosity of the MnO2 shell, as evidenced by
SEM analysis (Fig. S1a†). In the following step, dilute hydro-
chloric acid does not react with MnO2 but can penetrate
through the micropores of the MnO2 shell and react with the
MnCO3 cores. With the control of the HCl amount and the
reaction time, the MnCO3 core is partially etched and the yolk-
structured MnCO3 (core)–MnO2 (shell) microspheres are
generated (Fig. S1b†). Then, these spheres are heat-treated in air
at a high enough temperature (530 ꢀC) to convert both the
MnCO3 and MnO2 into Mn2O3. Thus, yolk-structured Mn2O3
spheres can be produced. Note that the Mn2O3 shells should be
Fig. 1 X-Ray diffraction patterns of (a) MnCO3 microspheres, (b) MnCO3–MnO2
core–shell structured precursor for LMO-Y, (c) hollow MnO2 precursor for LMO-H,
with micropores (Fig. S1c†). In the last step, yolk-structured (d) LMO-S, (e) LMO-Y, and (f) LMO-H.
862 | J. Mater. Chem. A, 2013, 1, 860–867
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 2 SEM images of microspheres of (a) MnCO3, (b) MnCO3 (core)–MnO2 (shell), (c) yolk-structured MnCO3 (core)–MnO2 (shell), (d) LMO-S, (e) LMO-Y, and (f) LMO-H.
The TEM images of (d) LMO-S, (e) LMO-Y and (f) LMO-H are also given as insets.
quickly with increasing the discharge rate from 0.2 C to 10 C.
Moreover, the capacity retention of LMO-S falls to 88% while
those of LMO-Y and LMO-H are both above 96% aer 100 cycles
at 25 ꢀC at 2 C rate (Fig. 3b). When the rate increases to 10 C, the
rst discharge capacities of LMO-S, LMO-Y and LMO-H are 73.4,
89.7 and 94.7 mA h gꢁ1, with the capacity retention of 76.8%,
95.3% and 87.9%, respectively (Fig. 3c).
In Fig. 3d, the discharge capacities as a function of discharge
rates of 0.2 C, 1 C, 2 C, 5 C and 10 C are compared. It obviously
reveals that the discharge capacities and cycle stability of
3.3 Electrochemical properties of solid, hollow and yolk-
structured LiMn2O4 powders
The cycling stabilities of the samples at room temperature are
tested by galvanostatically charging/discharging the cells under
different rates. Fig. 3a–c reveal the cycle performance of the
three LiMn2O4 samples at 25 ꢀC at 0.2 C, 2 C and 10 C between
3.2 and 4.35 V. The rst discharge capacities of LMO-S, LMO-Y
and LMO-H are 132.4, 128.9 and 130.2 mA h gꢁ1 at 0.2 C rate,
with the capacity retention of 92.9%, 98.2% and 96.2% aer 100
cycles at 25 ꢀC. The discharge capacity of LMO-S decreases
Fig. 3 Cycling performance of the LiMn2O4 samples at the rate of (a) 0.2 C, (b) 2 C, and (c) 10 C at 25 ꢀC and (d) plots of discharge capacity values of the LiMn2O4
samples at different rates from 0.2 C to 10 C.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 860–867 | 863

View Article Online
Journal of Materials Chemistry A
Paper
LMO-Y and LMO-H are quite similar. However, LMO-S suffers (Scheme 2). Thus, it can be derived that BETLMO-Y > BETLMO-H
from a substantial decrease with increasing the rate, which is only when r > 1.5d. As revealed by the TEM images (Fig. 2e), for
quite different from the other two samples. It is obvious that LMO-Y synthesized in this study, d z 100 nm and r z 350 nm,
both LMO-Y and LMO-H have a better rate capability than LMO- which satisfy the relationship r > 1.5d. Therefore, the specic
S. Moreover, LMO-Y shows a remarkable high capacity reten- surface area of LMO-Y should be greater than LMO-H. This is
ꢀ
tion, which still remains above 95% aer 100 cycles at 25 C, conrmed by the BET results (Table 1).
though the charge–discharge rate increases from 0.2 C to 10 C.
Fig. 5 reveals the cycle performance of the three LiMn2O4
ꢀ
The excellent high rate capability of LMO-Y may be due to its samples at 55 C at 1 C rate between 3.2 and 4.35 V. The rst-
specially designed yolk-structure. As shown in the TEM images cycle discharge capacities are 132.2, 130.5 and 131.5 mA h gꢁ1
,
of LMO-Y (Fig. 4), its shell consists of nano-particles compacted respectively. Aer 300 cycles, the retention rates of capacities
to form three-dimensional channels, which can signicantly are 67.6%, 86.6% and 52.1%, respectively. It is worth noting
facilitate the lithium ion diffusion and lead to the measured that the discharge capacity of LMO-Y still remains as 112.9 mA h
excellent rate capability. This can also be supported by their gꢁ1 aer 300 cycles at 1 C and 55 ꢀC. The charge and discharge
ꢀ
BET surface area measurements.
curves of LMO-S, LMO-Y and LMO-H at 1 C and 55 C are also
In order to increase the specic surface area and make sure shown in Fig. 5b–d. With increasing the cycle number, LMO-S
that the specic surface area of LMO-Y is greater than that of and LMO-H not only have a severe capacity fading (Fig. 5a), but
LMO-H, we should also control the core diameter and shell also suffer from an obvious polarization increase (Fig. 5b and
thickness of LMO-Y. According to the geometric relationship, d). But the yolk-structured LiMn2O4 (LMO-Y) shows a slight
their specic surface areas can be calculated with following polarization and better capacity.
equations:
In order to nd the reason for the difference in capacity
fading, we make a careful analysis of Fig. 5a. It is noticed that
the slopes of the LMO-S and LMO-Y curves maintain a stable
trend, while the LMO-H curve has an obvious drop at around
the 140th cycle. A possible reason might be the collapse of the
hollow microspheres in LMO-H in the process of charging and
discharging, leading to such a rapid capacity drop. The SEM
images of the LiMn2O4 electrodes aer 100 and 300 cycles at
r 3R2d þ d3 ꢁ 3Rd2 þ R3
BETLMO-Y ¼ $
(1)
(2)
3
2R2 þ d2 þ r2 ꢁ 2Rd
r 3R2d þ d3 ꢁ 3Rd2
BETLMO-H ¼ $
3
2R2 þ d2 ꢁ 2Rd
where r is the theoretical density of LiMn2O4, R the particle
radius, r the core radius of LMO-Y, and d the shell thickness
ꢀ
55 C at 1 C rate conrm this speculation.
3.4 Reasons for the superiority of the yolk-structured
LiMn2O4
As shown in Fig. 6a–c, although there are some small holes on
the surfaces of minority LMO-Y and LMO-H microspheres, all of
the three kinds of LiMn2O4 still maintain clear outlines of the
uniform spherical morphology. Aer 300 cycles (Fig. 6d–f), only
LMO-Y remains with an integrated spherical morphology
(Fig. 6e), while the surface of LMO-S becomes rougher and is
composed of closely packed nano-particles with the sizes below
100 nm (Fig. 6d). As speculated above, LMO-H spheres have
completely collapsed (Fig. 6f). It can be seen that the LMO-H
electrode contains dispersed particles with a particle size below
100 nm. This particle collapse may result in a signicant
increase in the specic surface area, which is conrmed by the
results of BET analysis (Table 2). The specic surface area of
LMO-Y is higher than that of LMO-H aer 100 cycles at 55 ꢀC at
Fig. 4 TEM (a) and SEM (b) images of LMO-Y with a high magnification.
Table 1 BET results of the precursors and the corresponding LiMn2O4 samples
Mn2O3
LiMn2O4
Samples
Precursor (m2 gꢁ1
)
(m2 gꢁ1
)
(m2 gꢁ1
)
LMO-S
LMO-Y
101.49 (MnCO3)
171.17
20.98
36.35
12.23
22.03
(yolk MnCO3 (core)–
MnO2 (shell))
148.71
LMO-H
31.09
19.66
Scheme 2 Schematic illustration of the geometry of a yolk-structured LiMn2O4
(hollow MnO2)
sphere.
864 | J. Mater. Chem. A, 2013, 1, 860–867
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 5 (a) Cycling performance of the LiMn2O4 samples at the rate of 1 C at 55 ꢀC; the charge and discharge curves of (b) LMO-S; (c) LMO-Y; (d) LMO-H after different
cycles at 55 ꢀC at 1 C rate.
Fig. 6 SEM images of (a) LMO-S, (b) LMO-Y, (c) LMO-H after 100 cycles at the rate of 1 C at 55 ꢀC; (d) LMO-S, (e) LMO-Y, and (f) LMO-H after 300 cycles at the rate of 1 C
at 55 ꢀC.
1 C, which is similar to the BET results before electrochemical
cycling (Table 1). But aer 300 cycles under the same conditions
the surface area of LMO-H increases to 29.71 m2 gꢁ1, which is
even greater than that of the cycled LMO-Y (25.13 m2 gꢁ1).
The cyclic voltammograms of the LMO-Y/Li cell aer being
cycled at 1 C rate at 55 ꢀC for different cycles in a potential range
of 3.2–4.35 V are shown in Fig. 7. No obvious difference can be
Table 2 BET results of the LiMn2O4 samples after different cycles at 55 ꢀC at 1 C
Samples
100th cycle (m2 gꢁ1
)
300th cycle (m2 gꢁ1
)
LMO-S
LMO-Y
LMO-H
13.69
23.47
21.88
15.94
25.13
29.71
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 860–867 | 865

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 9 Electrochemical impedance spectra of LiMn2O4/Li cells after (a) 5 cycles
and (b) 300 cycles at 55 ꢀC at 1 C rate. The open circuit voltages of all the cells
were 4.2 V.
transport, leading to increased polarization. If the decomposi-
tion products form a surface lm on the particles of electrodes,
the electrode with a higher specic surface area should suffer
from a higher degree of electrolyte decomposition and generate
more surface lms. Fig. 8 presents the IR spectra of LMO-S,
LMO-Y and LMO-H electrodes aer being cycled for different
times at 1 C and 55 ꢀC. The assignments for some major signals
are indicated in the spectra. Except for Li2CO3 which is formed
from the reaction of LiMn2O4 compound with CO2 in the air, the
electrodes also become covered with surface species whose
spectra very much resemble those of ROCOOLi species at about
1200 cmꢁ1. This species is believed to be formed in the ethylene
carbonate (EC)–dimethyl carbonate (DMC) solution on the
cathode side and is believed to precipitate on the surface of
cathode particles when the solution becomes saturated with
them aer few cycles. The intensity of this 1200 cmꢁ1 signal
reects the amount of the species and the surface lm. By
comparing the specic surface areas of the electrode aer
different cycles (Table 2), we can conclude that the amount of
decomposition product lms formed on the surface of the
LiMn2O4 electrode is proportional to their specic surface area.
We can further infer that the surface lm may hinder not only
the diffusion of lithium ions but also the charge-transfer
between the electrolyte and electrode. But the surface lm
uniformly deposited on the surface of the LiMn2O4 electrode
may also hinder the Mn dissolution, slow down the electrode
dissolution in the electrolyte and thus play a role of stabilizing
the electrode structure.
Fig. 7 Cyclic voltammograms (0.1 mV sꢁ1) of the LMO-Y electrode between 3.5
and 4.35 V vs. Li/Li+ after different cycles at the rate of 1 C at 55 ꢀC.
seen between the CV curves aer 1 cycle and aer 100 cycles.
Even aer 300 cycles, the CV curve has only some minor
changes. This indicates again that the LMO-Y electrode is very
superior in terms of reversibility for lithium ion intercalation/
deintercalation, even at elevated temperatures.
According to Aurbach et al.,31–33 electrolyte decomposition
may occur at high potentials. The decomposition products can
coat on the surface of the electrodes and hinder the lithium
Fig. 9 shows the electrochemical impedance spectra (EIS) of
ꢀ
the cells aer different cycles of cycling at 1 C and 55 C. They
were all measured at a same charge voltage plateau (4.2 V). It
can be seen clearly that aer 300 cycles the impedance of all the
three cells increases to different extents. The LMO-Y/Li cell has
the least rise in the impedance while LMO-H shows the greatest
increase. This result is consistent with the above structure
(SEM) and composition (IR) analyses. Therefore, LMO-Y has the
optimal microstructure to have the best electrochemical
properties.
To understand the superiority of LMO-Y, two factors seem to
be important to achieve the optimal electrochemical properties
of an electrode material. Firstly, a relatively high surface area is
needed to increase the contact area between the electrode and
electrolyte, leading to the minimization of internal resistance of
the cell. Secondly, good mechanical stability of the electrode is
necessary to withstand and endure the periodical volume
Fig. 8 FTIR spectra of the LiMn2O4 samples after (a) 100 cycles and (b) 300 cycles
at 55 ꢀC at 1 C rate. The 1200 cmꢁ1 region with larger magnification is also given
as an inset.
866 | J. Mater. Chem. A, 2013, 1, 860–867
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
change (expansion and contraction) of the electrode material
during the lithium intercalation and de-intercalation, particu-
larly over a large number of cycles. Among the LMO samples
with three different particle morphologies, LMO-Y reaches a
8 J. Cho and M. M. Thackeray, J. Electrochem. Soc., 1999, 146,
3577.
9 A. Yamada, M. Tanaka, K. Tanaka and K. Sekai, J. Power
Sources, 1999, 81, 73.
good balance between the high surface area and mechanical 10 X. F. Li, Y. L. Xu and C. L. Wang, J. Alloys Compd., 2009, 479,
stability so that it exhibits superior properties.
310.
11 A. Yamada, J. Solid State Chem., 1996, 122, 160.
12 X. F. Li and Y. L. Xu, Appl. Surf. Sci., 2007, 253, 8592.
13 I. Taniguchi, Ind. Eng. Chem. Res., 2005, 44, 6560.
14 M. W. Raja, S. Mahanty and R. N. Basu, J. Mater. Chem., 2009,
19, 6161.
15 M. Okubo, Y. Mizumo, H. Yamada, J. Kim, E. Hosono, Y. Cui
and D. K. Kim, Nano Lett., 2010, 10, 3852.
16 P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int.
Ed., 2008, 47, 2930.
17 Y. Wang and G. Z. Cao, Adv. Mater., 2008, 20, 2251.
18 J. Y. Luo, H. M. Xiong and Y. Y. Xia, J. Phys. Chem. C, 2008,
112, 12051.
19 Y. G. Guo, Y. S. Hu, W. Sigle and J. Maier, Adv. Mater., 2007,
19, 2087.
20 M. Okubo, Y. Mizuno, H. Yamada, J. Kim, E. Hosono,
H. Zhou, T. Kudo and I. Honma, ACS Nano, 2010, 4,
741.
21 J. R. Ying and C. Y. Jiang, J. Power Sources, 2004, 129,
264.
22 X. M. He, J. J. Li, Y. Cai, Y. Wang, J. R. Ying, C. Y. Jiang and
C. R. Wan, J. Power Sources, 2005, 150, 216.
23 J. Luo, L. Cheng and Y. Xia, Electrochem. Commun., 2007, 9,
1404.
24 J. Cho, J. Mater. Chem., 2008, 18, 2257.
25 J. F. Qian, M. Zhou, Y. L. Cao, X. P. Ai and H. X. Yang, J. Phys.
Chem. C, 2010, 114, 3477.
26 J. Cao, Y. C. Zhu, K. Y. Bao, L. Shi, S. Z. Liu and Y. T. Qian, J.
Phys. Chem. C, 2009, 113, 17755.
27 J. B. Fei, Y. Cui, X. H. Yan, W. Qi, Y. Yang, K. W. Wang, Q. He
and J. B. Li, Adv. Mater., 2008, 20, 452.
28 J. Cao, Y. C. Zhu, L. Shin, L. L. Zhu, K. Y. Bao, S. Z. Liu and
Y. T. Qian, Eur. J. Inorg. Chem., 2010, 1172.
29 X. M. He, J. J. Li, Y. Cai, Y. W. Wang, J. R. Ying, C. Y. Jiang and
C. R. Wan, J. Solid State Electrochem., 2005, 9, 438.
30 X. L. Xiao, J. Lu and Y. D. Li, Nano Res., 2010, 3, 733.
31 D. Aurbach, M. D. Levi, H. Teller, B. Markovsky, G. Salitra,
U. Heider and L. Heider, J. Electrochem. Soc., 1998, 9, 3025.
32 J. S. Gnanaraj, V. G. Pol, A. Gedanken and D. Aurbach,
Electrochem. Commun., 2003, 5, 940.
4 Conclusions
The spinel LiMn2O4 powder composed of yolk-structured
microspheres has been successfully prepared by a specially
designed multi-step synthesis procedure involving precipita-
tion, controlled oxidation, etching and chemical lithiation. The
shell and core of the yolk are respectively made porous and
dense to achieve high reactivity, structure stability and tap
density of the LiMn2O4 powder. The results show that the yolk-
structured LiMn2O4 shows a remarkable high rate charge–
discharge capability both at room temperature and elevated
temperatures. It exhibits a discharge capacity of 112.94 mA h
gꢁ1 with a capacity retention of 86.6% at 1 C rate and 55 ꢀC aer
300 cycles. Compared to LiMn2O4 powders with either solid
spheres or hollow spheres, this unique yolk-structured LiMn2O4
is more stable during electrochemical cycling, especially at
elevated temperatures. This reveals its excellent potential in
applications as a cathode material for rechargeable lithium-ion
batteries.
Acknowledgements
This study was supported by the National Science Foundation of
China (grant no. 20971117, 10979049 and J1030412), Education
Department of Anhui Province (grant no. KJ2009A142) and
National
Students’
Innovative
Training
Program
(cxxj103582012058). We are also grateful to the Solar Energy
Operation Plan of Academia Sinica.
Notes and references
1 A. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon and
W. V. Schalkwiik, Nat. Mater., 2005, 5, 366.
2 M. S. Whittingham, Chem. Rev., 2004, 104, 4271.
3 M. Armand and J. M. Tarascon, Nature, 2008, 451, 652.
4 M. M. Thackeray, W. F. David, P. G. Bruce and
J. B. Goodenough, Mater. Res. Bull., 1983, 18, 481.
5 D. Guyomard and J. M. Tarascon, J. Electrochem. Soc., 1992,
139, 937.
33 O. Haik, N. Leifer, B. Markovsky, L. Larush, Y. Goffer,
G. Goobes and D. Aurbach, J. Electrochem. Soc., 2010, 157,
A1199.
6 Y. Y. Xia and M. Yoshio, J. Electrochem. Soc., 1997, 144, 4186.
7 S. H. Park, K. S. Park, Y. K. Sun and K. S. Nahm, J.
Electrochem. Soc., 2000, 147, 2166.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 860–867 | 867