Gel-combustion synthesis of Li1.2Mn0.4Co 0.4O2 composites with a high capacity and superior rate capability for lithium-ion batteries

Article abstract of DOI:10.1039/c3ta13920d

Owing to the merits of high capacity and low cost, Li-rich layered composite cathode materials have received extensive attention. Nevertheless, such materials always suffer from a poor rate capability, which seriously hinders their widespread practical applications. In this work, Li _1.2Mn_0.4Co_0.4O₂ composites were fabricated by a gel-combustion method, in which lithium carbonates formed by an in situ burning reaction were homogeneously mixed with metal oxides, leading to excellent electrochemical properties. The sintering temperature and time were optimized to 900 °C and 15 h. The samples prepared at optimum conditions exhibited a high discharge capacity and excellent rate capability. At a current density of 20 mA g^-1, the specific discharge capacity was 310.5 mA h g^-1 for the first cycle and the capacity retention was 75.8% after 30 cycles. When the current densities increase by 10 times to reach 200 mA g ^-1, the initial discharge capacity is still as high as 241.4 mA h g^-1, superior to that of 203 mA h g^-1 at the same current density reported previously by the oxalate-precursor method. Even when the current densities increase by 20 times, the capacity remained as high as 203.7 mA h g^-1, and the capacity retention was 79.4% over 30 cycles. The high discharge capacity and improved rate capability of the optimized sample were beneficial with a perfect layered structure with c/a >5.0, appropriate particle size of about 450 nm, high lithium ion diffusion coefficient around 1.42 × 10^-13 cm² s^-1, and the presence of a maximum content of Mn³⁺ (11.6%), as determined by XPS. The preparation method reported herein may provide hints for obtaining various advanced Li-rich layered composite materials for use in high-performance energy storage and conversion devices.

Full text of DOI:10.1039/c3ta13920d

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Gel-combustion synthesis of Li1.2Mn0.4Co0.4O2
composites with a high capacity and superior rate
capability for lithium-ion batteries
Cite this: J. Mater. Chem. A, 2014, 2,
1471
a
b
a
a
a
*
Chaochao Fu, Guangshe Li, Dong Luo, Jing Zheng and Liping Li
Owing to the merits of high capacity and low cost, Li-rich layered composite cathode materials have
received extensive attention. Nevertheless, such materials always suffer from a poor rate capability,
which seriously hinders their widespread practical applications. In this work, Li1.2Mn0.4Co0.4O2
composites were fabricated by a gel-combustion method, in which lithium carbonates formed by an in
situ burning reaction were homogeneously mixed with metal oxides, leading to excellent
electrochemical properties. The sintering temperature and time were optimized to 900 ꢀC and 15 h. The
samples prepared at optimum conditions exhibited a high discharge capacity and excellent rate
capability. At a current density of 20 mA gꢁ1, the specific discharge capacity was 310.5 mA h gꢁ1 for the
first cycle and the capacity retention was 75.8% after 30 cycles. When the current densities increase by
10 times to reach 200 mA gꢁ1, the initial discharge capacity is still as high as 241.4 mA h gꢁ1, superior to
that of 203 mA h gꢁ1 at the same current density reported previously by the oxalate-precursor method.
Even when the current densities increase by 20 times, the capacity remained as high as 203.7 mA h gꢁ1
,
and the capacity retention was 79.4% over 30 cycles. The high discharge capacity and improved rate
capability of the optimized sample were beneficial with a perfect layered structure with c/a >5.0,
appropriate particle size of about 450 nm, high lithium ion diffusion coefficient around 1.42 ꢂ 10ꢁ13 cm2
Received 29th September 2013
Accepted 7th November 2013
s
ꢁ1, and the presence of a maximum content of Mn3+ (11.6%), as determined by XPS. The preparation
DOI: 10.1039/c3ta13920d
method reported herein may provide hints for obtaining various advanced Li-rich layered composite
materials for use in high-performance energy storage and conversion devices.
www.rsc.org/MaterialsA
Nevertheless, these materials have to face the challenge of
greatly increasing their capacities.
1. Introduction
Structurally integrated Li2MnO3–LiMO2 (M ¼ Mn, Ni, Co)
To meet the increasing demanding requirements for the storage
and utilization of clean energy, Li-ion batteries have been
proposed for their lower cost, higher capacity, faster charging/
discharging rate, longer cycling life, and better environmental
benignancy.1–5 However, the presently dominant cathode
material LiCoO2 has a reversible capacity of 140 mA h gꢁ1, only
50% of the theoretical capacity, as limited by its structural
stability.6–8 This low capacity cannot meet the high energy
requirements for practical applications. Another demerit is the
use of cobalt, an expensive and toxic element. Therefore, there
is immense interest in developing alternative cathode mate-
composites could provide high reversible capacities > 250 mA h
gꢁ1 and a relatively lower cost than LiCoO2, and therefore have
been regarded as very promising cathode materials.20 However,
the introduction of an insulating Li2MnO3 component in the
composites with low electronic conductivity always causes
inferior rate capability,21,22 which hinders their practical appli-
cations. To enhance the rate capability, great efforts have been
carried over the past years, which include: (i) the partial
substitution of foreign atoms, such as Cr, Mo, Mg, Al, Ga or
Ti,23–26 which improves the structure stability and cycle and rate
performance. However, introducing guest atoms into the crystal
lattices of the Li-rich layered composite materials may also
cause a lowering of capacity, because dopants are usually elec-
trochemically inactive; (ii) surface coating with metal oxides
(Al2O3),27 phosphates (AlPO4),28 or carbon materials29 which
could improve the stability of electrodes and prevent dissolu-
tion of Mn and side reactions at the electrode/electrolyte
interface, thus improving the rate capability. However, to ach-
ieve perfect surface coatings is oen extremely difficult. Owing
to the signicant relationship between the electrochemical
rials. In this regard, layered structure LiNixCoyMnzO2 (x + y + z ¼
18,19
1),9–13 olivine LiFePO4,14–17 and spinel LiMn2O4
are the
alternatives with the dual advantages of lower cost and toxicity.
aKey Lab of Optoelectronic Materials Chemistry and Physics, Fujian Institute of
Research on the Structure of Matter, University of Chinese Academy of Sciences,
Fuzhou, 350002, P. R. China. E-mail: lipingli@irsm.ac.cn; Fax: +86-591-83702122
bState Key Laboratory of Structural Chemistry, Fujian Institute of Research on the
Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, P. R. China
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1471

View Article Online
Journal of Materials Chemistry A
Paper
performance and synthetic method, an optimized synthetic (ICP-AES). Thermo-gravimetric (TG) analysis and differential
method perhaps is a better and more direct route to promote thermal analysis (DTA) were carried out on a Netzsch Model
ꢁ1
ꢀ
rate performance.
STA449F3 thermal analyzer at a heating rate of 10 C min in
Current preparation methods to prepare the composites are owing air from room temperature to 1200 ꢀC. The valence
usually based on mixed hydroxide or carbonate co-precipitation states of transition metal ions in the samples were determined
with lithium sources, which can result in micron-sized large by X-ray photoelectron spectroscopy (XPS) on an ESCA-LAB
spherical aggregates, and thus impede Li-ion transport.30 On MKII apparatus performed with a monochromatic Al Ka X-ray
the other hand, when using these conventional methods, raw source. During XPS measurements, the base pressure of the
materials cannot mix uniformly at a molecular level. Particu- sample chamber was kept below 3.0 ꢂ 10ꢁ10 Mbar. Emission
larly, some unwanted metal-ions (e.g. Na+) are inevitably intro- lines were calibrated with the C1s signal at 284.6 eV.
duced.31,32 All these may give rise to disadvantages in enhancing
the rate performance.
2.3. Electrochemical measurements
In this work, we employed a novel gel-combustion method
The electrochemical behaviour of the samples was examined
and prepared Li-rich Li1.2Mn0.4Co0.4O2 composites. During the
using a CR-2025-type coin cell. The cells were composed of a
synthesis, precursors of lithium, manganese and cobalt were
cathode and a lithium foil anode separated by a polymer
mixed uniformly. The resulting Li1.2Mn0.4Co0.4O2 composites
separator (Celgard 2320). The cathode lm was fabricated from
a mixture of the synthesized samples, carbon black (Super P Li,
Timcal) and binder (polyvinylidene uoride (PVDF), Alfa Aesar)
in a weight ratio of 8 : 1 : 1 using N-methyl pyrrolidinone (NMP,
Alfa Aesar) as the solution. The resulting slurry was coated on Al
foil by a doctor blade technique and dried at 100 ꢀC for 10 h. The
showed a high capacity and superior rate capability. The
methodology reported here may provide hints to obtain various
advanced Li-rich composites for use in high-performance
energy storage and conversion devices.
2. Experimental section
2.1. Sample syntheses
cells were assembled in an argon-lled glove box (H2O and O2 <
1 ppm) using 1 M LiPF6 in EC : EMC : DMC (1 : 1 : 1 in volume)
as the electrolyte. The cells were charged and discharged gal-
vanostatically at different current densities using a battery
cycler (Neware Test System) in the voltage range of 2.0 and 4.6 V
at room temperature. The electrochemical impedance spec-
troscopy (EIS) was measured using an electrochemical work-
station (CHI660C) and the applied frequency was from 100 kHz
to 10 mHz.
Li1.2Mn0.4Co0.4O2 composites were synthesized using
mannitol assisted gel-combustion method. The preparation
procedure can be briey described as follows: rstly, 27.6 mmol
a
CH3COOLi$2H2O
(15%
excess
lithium),
8
mmol
Mn(CH3COO)2$4H2O, and 8 mmol Co(CH3COO)2$4H2O were
dissolved in 100 mL water. Secondly, 100 mL of 0.2 M mannitol
solution was dropped into the above mentioned metal acetate
solution under stirring. Thirdly, the pH of the mixed solution
was adjusted to 2 using concentrated nitric acid to form a red
transparent sol. Then, the obtained sol was evaporated over-
night to dryness at 80 ꢀC and was heated up to burn sponta-
neously with a ame to form a brown ash. Finally, the brown
ash was sintered at 700, 800, 900, and 1000 ꢀC for 15 h,
respectively, and then quenched to room temperature. The nal
products were named as T700, T800, T900, and T1000, respec-
tively. The calcinations for different periods of time, such as 5,
10, and 20 h (named as H5, H10, and H20) at 900 ꢀC were also
conducted to optimize the electrochemical performance. The
synthesis process is depicted in Fig. 1. In addition, for conve-
nience of the discussion, T900 was also named as H15.
3. Results and discussion
3.1. Gel-combustion synthesis of the Li1.2Mn0.4Co0.4O2
composites
Until now, layered Li-rich composites have been mainly
prepared by a conventional co-precipitation method, in which
hydroxides or carbonates precursors are not easily homoge-
neously mixed with lithium sources. As a result, micron-sized
spherical aggregates are always obtained, which would impede
Li-ion transport. In this work, we designed a novel mannitol
assisted gel-combustion method with an aim to overcome the
drawbacks of the conventional preparation methods.
Fig. 1 illustrates the synthetic process we have newly
designed for Li1.2Mn0.4Co0.4O2 composites. During the
synthesis process, mannitol plays dual important roles: (i) as a
2.2. Sample characterization
The phase structure and purity of the samples were character- complexing agent. The metal ions coordinated with the
ized by powder X-ray diffraction (XRD) on a Rigaku Miniex hydroxyl of mannitol may make the solution form a sol and
˚
apparatus (Cu Ka, l ¼ 1.5418 A). Lattice parameters of the ensure the uniform mixing of the metal ions at a molecular
samples were calculated by Rietveld renement using GSAS level. (ii) As a combustion agent. Heat from the redox reaction of
soware.33 KCl powder serves as an internal standard for peak mannitol and nitrate may make the gel ame. As shown in
positions calibration. The particle sizes and morphologies of Fig. 2a, the brown ash precursors obtained aer combustion at
the samples were observed by eld-emission scanning electron 200 ꢀC are composed of lithium carbonate (Li2CO3) and oxides
microscopy (SEM) (JEOL, model JSM-6700). Chemical compo- of manganese and cobalt. Thus, Li2CO3 obtained during the
sitions of the samples were quantitatively determined using in situ burning reactions could be uniformly mixed with metal
inductively coupled plasma atomic emission spectrometry oxides, which is benecial for achieving excellent
1472 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 1 Illustration of the preparation process of the Li1.2Mn0.4Co0.4O2 cathode material by the gel-combustion method.
performance.30,34–36 Therefore, using the synthetic process
designed in this work it is possible to achieve optimized
materials with superior electrochemical performance.
3.2. Optimized preparation conditions: impact of sintering
temperature
To optimize the sintering temperature, the TG and DTA of the
brown ash precursor were examined, as indicated in Fig. 2b.
The TG data showed that the mass loss process underwent three
steps. In the rst one, a continuous mass loss occurred below
200 ꢀC, which could be assigned to the evaporation of water.
The second one involved a rapid weight loss from 230 to 500 ꢀC,
accompanied by an exothermic peak on the DTA curve, indi-
cating the formation of a layered phase above 500 ꢀC. The third
one is characterized by a continuous slow mass loss from 500 to
1200 ꢀC, which could be ascribed to the decomposition of
Li2CO3 in the layered-phase formation process and a small
amount of evaporation of lithium. Therefore, we sintered the
ꢀ
precursors at 700, 800, 900, and 1000 C, respectively, to opti-
mize the synthetic conditions.
Fig. 2 (a) XRD pattern and (b) TG and DTA curves of the brown ash
precursor obtained from burning the gel.
In order to make clear the components of the obtained
samples at different temperatures, ICP-AES measurements were
conducted, as listed in Table 1. One can see that the molar
ratios of Li : Mn : Co were closer to the initial molar ratios.
The impacts of the sintering temperature on the structures
electrochemical properties. Our previous work34 has reported
that gas CO2 would be released when Li2CO3 reacts with tran- of the samples were monitored by XRD, as shown in Fig. 3. The
sition metal oxides to form the layered materials. The released strong diffraction peaks were all indexed well in a hexagonal
gas (CO2) reduces the oxygen partial pressure near the reactants a-NaFeO2 structure with the R-3m space group of layered
and results in the presence of Mn3+ in Li2MnO3ꢁd. The presence LiCoO2, while weak peaks attached onto the strong (003) peak at
of Mn3+ ions in the lithium-rich composites could improve the 2q ¼ 20–25ꢀ are indicative of the existence of the monoclinic
electronic conductivity and furthermore the electrochemical Li2MnO3 phase (C2/m) with a cation ordering between Li and
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1473

View Article Online
Journal of Materials Chemistry A
Paper
Table 1 Experimentally determined compositions of the samples and lattice parameters of the layered LiCoO2 component in the samples
Experimental metal-ion molar ratios
Lattice parameters
3
˚
a (A)
˚
c (A)
˚
Samples
Li
Mn
Co
c/a
V(A )
T700
T800
T900
T1000
1.167
1.173
1.141
1.143
0.411
0.416
0.420
0.421
0.421
0.411
0.440
0.436
2.8376(1)
2.8361(1)
2.83081(7)
2.82869(6)
14.159(1)
14.1658(8)
14.1663(7)
14.1656(5)
4.9897
4.9948
5.0043
5.0078
98.73(1)
98.676(6)
98.312(7)
98.161(5)
high crystallinity, good hexagonal ordering, and improved
layered characteristics. In short, Li1.2Mn0.4Co0.4O2 with an
improved layered structure was successfully synthesized by the
current gel-combustion method.
Fig. 4 displays the typical morphologies and particle size
distributions taken by SEM for the samples sintered at 700–
1000 ꢀC. It is seen that the particle sizes and particle size
distributions exhibited huge differences for all of the samples.
Samples T700 and T800 showed nearly the same particle size,
mainly distributed around 250 nm. Sample T900 synthesized at
Fig. 3 XRD patterns of the samples synthesized at different sintering
temperatures. Black asterisks (*) denote the internal standard KCl.
the transition metal layer.37–40 For sample T700, weak peaks
around the two theta of 32ꢀ are assigned to the residual Li2CO3.
With increasing the sintering temperature, the splitting of the
pair reections (006)/(012) and (018)/(110) became more
obvious, indicating the improvement of the layered
structure.41,42
Lattice parameters of the layered LiCoO2 for the samples
were determined using GSAS soware,33 as exhibited in Table 1.
It is well documented that the axis ratio of c/a is an indicator of
the hexagonal ordering.43 That is, the larger c/a, the better the
hexagonal ordering. In this regard, the axis ratio c/a was taken
as an important structural parameter to examine the impacts of
the sintering temperature on the lithium de-intercalation in the
lattice of the samples. As shown in Table 1, as the temperature
increased from 700 to 1000 ꢀC, the axis ratio of c/a gradually
increased from 4.9897 to 5.0078, which also indicates that the
layered structure was improved by elevating the temperature.
Meanwhile, the intensities of the diffraction peaks increased,
and their full width at half-maximum (FWHM) of the reection
peaks decreased, demonstrating the better crystallinity and the
Fig. 4 SEM images and particle size distributions of the samples
prepared at different temperatures: (a and b) T700, (c and d) T800,
larger particle sizes. Therefore, the as-prepared samples have (e and f) T900, and (g and h) T1000.
1474 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
900 ꢀC had a slightly bigger particle size, and the mean particle the elevated sintering temperature is benecial for activating
size was about 450 nm. Moreover, it has a smooth surface, the Li2MnO3 component and achieving a higher discharge
which indicates good crystallization. When the sintering capacity. The reasons for this observation may be attributed to
temperature increased to 1000 ꢀC, the particles grew into blocks the improved crystallinity, better layer structure, and increased
with a size distribution in the range of 1–3.5 mm, as shown in particle size. However, when the sintering temperature further
Fig. 4h. These results are compatible with the XRD analyses. It is increased to 1000 ꢀC, the electrochemical performance began to
thus concluded that the sintering temperature has a great decrease. The length of the plateau region, the discharge
inuence on particle size, and that a high sintering temperature capacity, and coulomb efficiency were only about 180 mA h gꢁ1
is benecial for the formation of bigger particle sizes.
155.4 mA h gꢁ1, and 55.2%, respectively. These results may be
,
The electrochemical performances for the samples sintered due to the oversized particles (about 2 mm), which result in
at different temperatures are compared in Fig. 5. Fig. 5a depicts longer lithium intercalation and de-intercalation paths.
the initial charge and discharge curves between 2.0 and 4.6 V for
The cycle performance of the samples sintered at different
the Li1.2Mn0.4Co0.4O2 at a current density of 20 mA gꢁ1. One can temperatures was further compared, as indicated in Fig. 5b. For
see two charging regions corresponding to different electro- samples T700 and T800, the discharge capacities quickly
chemical reactions. The rst region is ascribed to the oxidation decreased with an increase in the cycle number. Aer 30 gal-
of cobalt ions from Co3+ to Co4+ in the voltage region of 3.8 to vanostatic charge–discharge cycles, the discharge capacities
4.4 V, while the subsequent plateau region is assigned to the were only 57.6 and 106.2 mA h gꢁ1, respectively. For sample
irreversible removal of Li2O from the electrode.44–46 The capacity T900, the discharge capacity was still up to 234.1 mA h gꢁ1, and
of the sloping region for the samples prepared at different maintained 75.4% of its initial capacity. The sample T1000
temperatures is almost the same. However, there is a big exhibited a relatively high cycle capability, but with a lower
difference for the plateau region that comes from the activation discharge capacity. It is thus obvious that the sample T900
of Li2MnO3. With increasing the sintering temperature from displayed the most excellent electrochemical performance
700 to 900 ꢀC, the length of the plateau region gradually among all of the samples. Apparently, the sintering temperature
increased, and the sample T900 exhibited the longest length, dependence of the electrochemical performance observed in
which reached up to about 300 mA h gꢁ1. Meanwhile, the this work is totally different from that previously reported for
discharge capacities gradually increased from 151.3, 208.6, to Li1.2Mn0.54Co0.13Ni0.13O2 composites by Croguennec et al.47 For
310.5 mA h gꢁ1, and the coulomb efficiency also gradually the latter case, the sintering temperatures are announced to
increased from 54.7, 58.3, to 75.8%. These results indicate that have less impact on the electrochemical performance of the
composites Li1.2Mn0.54Co0.13Ni0.13O2. Therefore, the chemical
compositions and choice of chelating agent (like organic acid or
alcohol) could alter the burning temperatures of the precursors
and also the microstructure and properties of the desired
samples (e.g., particle size, valence state of the transition
metals, lattice parameters, and electrochemical performance).
The results mentioned above reveal that the sintering
temperature could be used to greatly tune the crystallinity,
structure, particle size of the composites, and the length of the
plateau region in the rst charging process. A low sintering
temperature gives rise to smaller particle sizes, but lower crys-
tallinity and poor layered structures, leading to a poor cycle
performance and lower discharge capacity. Higher sintering
temperatures result in higher crystallinity, a better layered
structure, but larger particle size, also leading to poor electro-
chemical performance. Apparently, 900 ꢀC is the optimal sin-
tering temperature. Therefore, in the following, all of the
samples were prepared at 900 ꢀC with different periods of time
to obtain the optimum electrochemical performance.
3.3. Optimized preparation conditions: impact of sintering
time
3.3.1. Electrochemical performance of the samples
prepared with different sintering times. Fig. 6a shows the initial
ꢀ
charge–discharge curves of the samples prepared at 900 C for
5, 10, 15, and 20 h at a current density of 20 mA gꢁ1 in a voltage
window of 2.0–4.6 V at room temperature. Upon the rst charge
Fig. 5 (a) Initial charge–discharge curves and (b) cycle performance at
a current of 20 mA gꢁ1 between 2.0 and 4.6 V for the samples
synthesized at 700, 800, 900, and 1000 ꢀC.
to 4.6 V, the electrochemical reaction of lithium with the
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1475

View Article Online
Journal of Materials Chemistry A
Paper
sintering time increased to 20 h, the discharge capacities were
the lowest among all of the samples. That is to say that the
sample H15 has the best capacity performance. Aer 30 cycles,
the discharge capacities and capacity retention ratio of the
sample H5 were 143.6 mA h gꢁ1 and 60.7%, while those of the
sample H10 were 211.9 mA h gꢁ1 and 77.9%. For sample H15,
aer 20 cycles, the discharge capacity of the Li1.2Mn0.4Co0.4O2
electrode reached up to 250.4 mA h gꢁ1, larger than the retained
capacity of 195 and 235 mA h gꢁ1, respectively, reported previ-
ously.31,48 Moreover, aer 30 cycles, the discharge capacity still
remained at 234.1 mA h gꢁ1, and the capacity retention ratio
was 75.4%. In addition, the discharge capacities and capacity
retention ratio of sample H20 was 109.8 mA h gꢁ1 and 55%. It is
obvious that the sintering time has a huge inuence on the
cycling and capacity performance.
In order to investigate the electrochemical performance of
the samples H5, H10, H15, and H20 at high current densities,
the charge and discharge proles were conducted at room
temperature in a voltage range between 2.0 and 4.6 V. The rst
charge–discharge curves of all the samples at different current
densities are exhibited in Fig. 7. As mentioned above, all the
charge curves consisted of a sloping region and a plateau region
at approximately 3.8–4.4 V and 4.4–4.6 V, respectively. Both
voltage plateaus shied towards higher potentials with an
increase in the current densities for all of the samples, This
indicates that the polarization in the electrodes increased with
the increasing current densities, which may be associated with
the poor conductivity of the lithium-rich composites.21,22
In addition, with increasing the current density from 60 to
Fig. 6 (a) Initial charge–discharge curves and (b) cycling performance
at a current density of 20 mA gꢁ1 in a voltage range from 2.0 to 4.6 V
for the samples synthesized at different sintering times at 900 ꢀC.
Li1.2Mn0.4Co0.4O2 electrode occurs in two distinct steps as
follows: the rst step occurs at a voltage of <4.4 V vs. Li/Li+, 400 mA gꢁ1, the discharge capacities of all of the samples
which corresponds to the lithium extraction from LiCoO2 decreased. As seen from Fig. 7a, the discharge capacity for
component with the concomitant oxidation of Co3+ to Co4+. The sample H5 decreased from 229.7 to 140.7 mA h gꢁ1. The
second step corresponds to the removal of lithium from the discharge capacities of the sample H10 were 227.4, 218.3, 189.4,
Li2MnO3 component accompanied by oxygen evolution when and 194.7 mA h gꢁ1, respectively, at current densities of 60, 100,
charged above 4.4 V vs. Li/Li+. During the oxygen loss plateau, 200, and 400 mA gꢁ1, (Fig. 7b). These discharge capacities are
the Li2MnO3 component was activated to form the MnO2-like the second largest. Sample H15 exhibited the largest discharge
component44–46 and thus the material could deliver a high capacity among all four of the samples. The rst discharge
discharge capacity during the subsequent discharge process.
capacities were 258.1, 245.7, 241.4, and 203.7 mA h gꢁ1 at
As indicated in Fig. 6a, the obtained samples H5, H10, H15, current densities of 60, 100, 200, and 400 mA gꢁ1, respectively.
and H20 cathodes delivered discharge capacities of 230.1, 271.9, In our previous work,31 the Li1.2Mn0.4Co0.4O2 composite
310.5, and 199.6 mA h gꢁ1, respectively, with an initial prepared by an oxalate-precursor method exhibited a rst
coulombic efficiency of 64.9%, 69.7%, 75.8%, and 51.6%, discharge capacity of 203 mA h gꢁ1 at a current density of 200
calculated from the rst charge–discharge capacity values at a mA gꢁ1. By the current gel-combustion method, the rst
specic current density of 20 mA gꢁ1. The discharge capacity of discharge capacity was increased to 241.4 mA gꢁ1 at the same
the sample H15 is higher than the previous reports.31,48,49 It is current density. Both values are higher than the reference
clear that, when compared to the samples H5 and H10, the reports.48 However, with the sintering time increased to 20 h,
sample H15 sintered at 900 ꢀC for 15 h had a larger charge the discharge capacities of the sample H20 shown in Fig. 7d
plateau above 4.4 V and further exhibited a relatively high were only 159.1, 142.6, 130.8, and 109.7 mA h gꢁ1, respectively.
discharge capacity. This observation indicates that a longer The results indicate that sample H15 had a fairly high discharge
sintering time is benecial for activation of the Li2MnO3 capacity at high current densities and superior rate
component. But when the sintering time increased to 20 h, the performance.
length of charge plateau above 4.4 V began to decline, and the
The cycle and rate performance of the samples H5, H10, H15,
coulombic efficiency was the lowest among all of the samples. and H20 at high current densities were also investigated, as
Fig. 6b compares the cycling performance of all of the shown in Fig. 8. At the same current density of 60 mA gꢁ1, the
samples between 2.0 and 4.6 V at a current density of 20 mA gꢁ1
.
discharge capacities were 167.3, 202.3, 190.6, and 138.3 mA h
It is clear that the discharge capacities of the 30 cycles gradually gꢁ1 aer 30 cycles, for sample H5, H10, H15 and H20, and their
increased with the sintering time from 5 to 15 h. When the corresponding capacity retention ratios relative to the highest
1476 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 7 Initial charge–discharge curves at different currents between 2.0 and 4.6 V for the samples: (a) H5, (b) H10, (c) H15, and (d) H20
synthesized using different times at 900 ꢀC.
capacity among 30 cycles were 83.4%, 85.4%, 73.8%, and 80.6%, ratio of 84.7%, 87.4%, 82.7%, and 85.6%, respectively.
respectively. When the current densities increased to 200 mA Furthermore, when the current densities increased to
gꢁ1, the discharge capacities of all of the samples were 140.0, 400 mA gꢁ1, the discharge capacities of all of the samples still
177.6, 204.1, and 125.3 mA h gꢁ1 with the capacity retention remained as 123.0, 168.1, 173.0, and 100.8 mA h gꢁ1, with the
Fig. 8 Cycle performance of the samples: (a) H5, (b) H10, (c) H15, and (d) H20 at current densities of 60, 100, 200, and 400 mA gꢁ1
.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1477

View Article Online
Journal of Materials Chemistry A
Paper
capacity retention ratios of 83.7%, 83.8%, 79.4%, and 83.4%,
respectively. It can be seen from these results that sample H10
has the best cycling performance among all of the samples.
Compared to the samples H5, H10, and H20, the sample H15
has a superior rate capability, as evidenced by an almost over-
lapping of the curves of the discharge capacity vs. the cycle
number curves. All of this demonstrates that sample H15
exhibited an optimum electrochemical performance, as repre-
sented by having the largest capacity, best rate capability and
relatively high cycling performance.
In short, with prolonging the sintering time, the electro-
chemical performances of the composites started to show an
optimum. When the sintering time was prolonged from 5 to
15 h, the electrochemical performance of the composites
exhibited a huge difference. Sample H5 had a low discharge
capacity at different current densities, and sample H10 showed
the best cycle performance, while sample H15 had the highest
discharge capacity, the best rate performance, and relatively
high cycle performance. When the sintering time was extended
to 20 h, the electrochemical performances began to decrease to
show the lowest among all of the samples.
Then, an important question appears: what are the primary
reasons for the variations of electrochemical performance with
sintering time? To answer this question, several factors that
may contribute to the electrochemical performance, such as
chemical composition and structure, particle size, electro-
chemical redox reactions, reaction kinetics and diffusion coef-
cient of the lithium ion have to be taken into account.
3.3.2. Reasons for the differences in the electrochemical
performance. In order to understand the differences of the
Fig. 9 XRD patterns of the samples synthesized using different times
(5, 10, 15, and 20 h) at 900 ꢀC. Black asterisk symbol (*) denotes the
internal standard of KCl.
The particle size of the cathode materials may impact on the
electrochemical performance among the samples H5, H10, electrochemical performance. As the particle size decreases, the
H15, and H20, the phase structure analyzed by XRD renement transport path of the lithium ion becomes shorter, which can
using GSAS soware, particle size estimated by SEM, electro- improve the kinetic properties of the electrodes to some extent,
chemical redox potential determined from differential capacity leading to a higher rate capability.11 Fig. 10 shows the
vs. voltage (dQ/dV) curves, as well as the diffusion coefficient of morphologies and particle size distributions obtained by SEM
ꢀ
lithium ion based on electrochemical impedance spectroscopy observations on samples prepared at 900 C for 5, 10, 15, and
(EIS) were systematically examined.
20 h. It is seen that the constituent particles for all of the
Due to the importance of the relationship between the samples were approximately round and obviously aggregated.
structure and electrochemical performance of the electrode As the sintering time increased, the particle sizes and particle
materials for lithium batteries, it is necessary to rstly under- size distributions for the samples H5, H10, and H15 was closer,
stand the structure of the Li1.2Mn0.4Co0.4O2 composites. Fig. 9 and their mean particle sizes were 390, 430, and 450 nm,
shows XRD patterns of the samples sintered at 900 ꢀC for 5, 10, respectively. However, when the sintering time increased to
15, and 20 h. These patterns showed similar characteristics to 20 h, the sample H20 had a bigger particle size, and its mean
those at different temperatures in Fig. 3, which indicates the particle size was increased up to 830 nm. These observations
formation of Li2MnO3 and LiCoO2 composites. Through XRD suggest that the particle size effect on the electrochemical
renement using GSAS soware,33 lattice parameters for the performance for samples H5, H10, and H15 could be ignored,
layered R-3m structures were obtained, as shown in Table 2. The while the biggest particle size for sample H20 among all the
data showed that the values of a and c decreased, while the axis samples should be the primary reason for its worse electro-
ratio of c/a was not obviously changed with increasing the sin- chemical performance.
tering-time. Moreover, the values of c/a for all of the samples are
To recognize the electrochemical redox reactions which
greater than 5.0,31,43 indicating that all of the samples had a might occur during the charge and discharge processes,
good layered structure. Therefore, the structure factor can be differential capacity curves (dQ/dV vs. V) of different cycles for
dismissed. The chemical compositions for all of the samples all of the samples sintered at different times were plotted, as
were detected by the ICP-AES technique. As shown in Table 2, shown in Fig. 11. In the rst cycle, the oxidation peaks at 3.9–
the molar ratios of Li and Mn to Co were closer to each other 4.0 V are attributed to the oxidation from Co3+ to Co3.6+, and
within an error range, approaching the designed ratios. Hence, that at about 4.36 V is assigned to the oxidation process of
the inuence of the chemical compositions were also neglected. Co3.6+/Co4+, which is consistent with the sloping region in the
1478 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Table 2 Experimentally determined compositions for the samples H5, H10, H15, and H20, and lattice parameters of the layered LiCoO2
component in the Li1.2Mn0.4Co0.4O2 composites for samples H5, H10, H15, and H20
Experimental metal-ion molar ratios
Lattice parameters
3
˚
a (A)
˚
c (A)
˚
Samples
Li
Mn
Co
c/a
V (A )
H5
1.152
1.159
1.141
1.163
0.424
0.419
0.420
0.414
0.425
0.421
0.440
0.423
2.83802(6)
2.83525(7)
2.83081(7)
2.83021(7)
14.2078(5)
14.1849(7)
14.1663(7)
14.1629(6)
5.0062
5.0031
5.0043
5.0042
99.103(5)
98.751(6)
98.312(7)
98.247(5)
H10
H15
H20
plateau above 4.4 V which predominantly corresponds to the
highly irreversible activation process of stripping Li2O off the
Li2MnO3 component and forming an active MnO2-like compo-
nent.44–46 Because of the irreversibility of the activation reaction,
no corresponding reduction peak was observed in the initial
discharge process, but it induces a new peak at 3.43–3.47 V,
which stems from the reduction reaction of Mn4+ to Mn3+.31,49 In
the second cycle, peaks at about 4.6 V were observed for samples
H5 and H10, which stem from the further activation process of
the residual Li2MnO3 component. With the cycles increasing,
two new conspicuous oxidation peaks were observed around
3.02/3.04 and 3.29/3.28 V in the charging process for samples
H10 and H15. Both two oxidation peaks can be ascribed to the
Li+ de-intercalation from the spinel-like structure (in 3.02/3.04 V
region) and layered LiMnO2 structure (in 3.29/3.28 V region),
respectively.31 Their corresponding reduction peaks appeared at
around 2.86 and 3.0 V. Furthermore, when compared to the
sample H10, these two redox couples are more obvious for the
sample H15. Comparing to sample H5 that only showed one
new conspicuous oxidation peaks around 3.05 V, there are not
any new conspicuous oxidation peak appearing for sample H20.
These results demonstrate the existence of trivalent manganese
ions in sample H5, H10, H15, and H20, and the content of
trivalent manganese ions among these four samples increases
in the sequence: H20 < H5 < H10 < H15 aer some cycles.
According to the reports in literature,30,34–36 some Mn3+ ions in
the lithium-rich composites could improve the electronic
conductivity and furthermore the electrochemical performance.
Therefore, the content of Mn3+ could be one primary reason for
the variation of the discharge capacities.
The different particle size and content of Mn3+ for the
samples H5, H10, H15 and H20 should be associated with the
sintering time. It is well known that 900 ꢀC is the decomposition
temperature of Li2CO3, which is higher than its melting point.34
Therefore, in the initial stage of Li2CO3 decomposition, with the
short sintering time of 5 h, the generated CO2 gradually
Fig. 10 SEM images and particle size distributions of the samples: (a
and b) H5, (c and d) H10, (e and f) H15, (g and h) H20 prepared using
different times at 900 ꢀC.
rst charge curve.31,44,45,50,51 The corresponding reduction peak removes the air from the surfaces of the reactant particles and
of Co4+ to Co3+ is also observed at 3.9 V in the discharge process. reduces the oxygen partial pressure, resulting in the slight
The whole redox couple shows that the reaction of Co3+/Co4+ is content of Mn3+. In the second stage, as the sintering time
reversible. Among the four samples, the peaks derived from the increases (i.e. 10 and 15 h), air near the surface of the reactant
redox couple of Co3+/Co4+ for sample H5 is sharper than those particles was further eliminated by the released CO2 gas,
of the other samples, which indicates that the contribution to resulting in a bigger content of Mn3+. Finally, when the sinter-
the total capacity from LiCoO2 component for sample H5 is ing time increased to 20 h, since Li2CO3 is almost decomposed
greater than others. The very strong peak observed at 4.45– completely, the released CO2 gas gradually decreased and the
4.47 V for all of the four samples matched with the voltage oxygen partial pressure increased. Therefore, the amounts of
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1479

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 11 Differential capacity vs. voltage (dQ/dV) curves corresponding to the charge–discharge curves of different cycles at a current density of
20 mA gꢁ1 for the samples: (a) H5, (b) H10, (c) H15, (d) H20.
Mn3+ would be re-oxidized to Mn4+. In the rst two stages, some upon valence change; and (ii) the splitting of the double peaks
amounts of Li2CO3 were melted and the reaction between the in Mn 3s is 4.5–4.7 eV for Mn4+ ions, and the value increases
liquid and solid occurred, leading to mean particle sizes that obviously to 5.2–5.5 eV when the valence state of Mn ions
were almost the same for the samples H5, H10, and H15. decreases to +3.56 The Mn 3s spectra are illustrated in Fig. 12(i–
However, when the sintering time increased to 20 h, all the l). The main broad peak at 84.4 eV is asymmetric at the lower
reactions mainly preceded in solid–solid forms, which gave rise binding energy side, which could be well tted in two peaks
to the apparent particle growth.
using mixed Gaussian–Lorentzian proles. Similarly, its
In order to prove the above hypothesis, XPS spectra of the multiple splitting peak at a higher binding energy is also
samples were measured to examine the valence states of Mn decomposed into two peaks. During the data t, two constraints
and Co ions. As shown in Fig. 12(a–d), four signals that include have to be imposed: (i) the splitting of two doublets were set at
two spin–orbit splitting peaks and two corresponding satellite about 4.6 eV for Mn4+ and 5.5 eV for Mn3+, respectively, and (ii)
peaks were detected for the Co 2p core level. Co 2p3/2 binding FWHM values of photoelectron peaks were smaller than 2.6 eV.
energies for samples H5, H10, H15, and H20 were around 780 From the tting results, one can see that the relative content of
eV, and their spin–orbit splitting values were all about 15 eV. Mn3+ varied in a range from 5.0 to 11.6% in the sequence: H20 <
Moreover, all the Co 2p core levels exhibited a sharp 2p3/2 peak: H5 < H10 < H15, as indicated in Fig. 12(i–l).
the full-width at half-maximum (FWHM) is only about 1.9 eV,
Electrochemical impedance spectroscopy (EIS) represents an
and the binding energy difference between 2p3/2 and its satellite important technique for evaluating the interfacial electro-
is about 9.6 eV. All of this indicates that the cobalt ions for all chemistry, and reaction kinetics and diffusion coefficient of the
the samples are in an oxidation state of +3 with a low-spin lithium ion in Li-ion battery materials.31,58 Impedance spectra
state.52–54
for the samples prepared at 900 ꢀC with different sintering
Fig. 12(e–h) shows the XPS data for Mn 2p, in which two periods of time were measured on the Li1.2Mn0.4Co0.4O2/Li cells
broad peaks with FWHM > 3 eV were observed. Both peaks are which were charged to 4.6 V in initial charge. Fig. 13 shows the
assigned to Mn 2p3/2 and 2p1/2. The binding energy difference Nyquist plots of Li1.2Mn0.4Co0.4O2 electrode test from 100 KHz
between 2p3/2 and 2p1/2 was about 11.5 eV, which indicates that to 10 mHz, in which Zi and Zr denote the imaginary and real part
most of the Mn ions are present in +4. The broadened FWHM of impedances, respectively. Corresponding proles of Zr vs.
observed also means the occurrence of mixed valence Mn3+ and uꢁ1/2 in the frequency region of 0.1–0.01 Hz are used to calcu-
Mn4+.55–57 To determine the relative content of Mn3+, Mn 3s late the Warburg factor. The impedance spectra were analyzed
spectra were studied, which is based on the following consid- by Zswinpwin soware and the data tted to the equivalent
erations: (i) Mn 3s electrons belong to the outer level ones, electrical circuit is shown in the insert of Fig. 13a. Here, Re
which are sensitive to the chemical environment, especially represents the internal resistance of the cell, Rsf and Csl are the
1480 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 12 XPS data of (a–d) Co 2p, (e–h) Mn 2p, and (i–l) Mn 3s for sample H5, H10, H15, and H20, respectively. Symbol “S” represents the satellite
peaks for the Co 2p levels.
resistance and capacitance of the SEI lm, the formation of
From eqn (2), the Warburg factor (s) can be obtained from
which involves a series of spontaneous reactions between the the linear tting of Zr vs. uꢁ1/2, as shown in Fig. 13b. The
cathode active materials and the electrolyte solvents. Rct and Cdl calculated results of lithium diffusion coefficient for sample H5,
(in Fig. 13a) are the charge transfer resistance due to the lithium H10, H15, and H20 were exhibited in Table 3.
intercalation/de-intercalation process and double layer capaci-
As indicated in Fig. 13a, all the Nyquist plots exhibited two
tance, and W describes the Warburg impedance which is semicircles in the high and middle frequency region and a slope
directly related to the solid state diffusion of lithium ion inside in the low frequency region. The values of Re, Rsf, Rct, and Rtotal
the active particles, signied by the straight sloping line at the (total resistances) for samples H5, H10, H15, and H20 are given
low frequency region.
in Table 3. It can be seen that sample H5 had the highest Rtotal
The diffusion coefficient of lithium ion (DLi+) is calculated and Rct, while sample H10 had the lowest Rtotal and Rct, which
from the straight sloping line at low frequency region according indicates that the sample H10 had the maximum electronic
to the following equation,31,34
conductivity, accounting for it having the best cycling perfor-
mance and relatively high rate performance. Sample H15 had
the maximum lithium diffusion coefficient and second smallest
Rct value, leading to the best electrochemical performance when
comparing among all of the samples. In addition, sample H20
had a relatively low lithium diffusion coefficient and high Rtotal
and Rct, explaining its inferior electrochemical performance.
In addition, all four samples exhibited a capacity fading with
an increase of the cycle number, which might be due to the
following reasons: (1) dissolution of the cobalt59 and manga-
nese, in which dissolution of the Mn happens in
+
DLi ¼ 0.5R2T2/n4A2F4C2s2
(1)
where R is the ideal gas content, T is the absolute temperature, n
is the number of electron(s) per molecule oxidized, A is the
surface area of the electrode, F is the Faraday constant, C is the
concentration of Li+ in the cathode, and s is the Warburg factor
which has the relationship with Zr as in the following equation,
Zr ¼ Re + Rsf + Rct + suꢁ1/2
(2)
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1481

View Article Online
Journal of Materials Chemistry A
Paper
Namely, Li2CO3 obtained during the in situ burning reaction
was homogeneously mixed with metal oxides, which is bene-
cial for achieving excellent electrochemical properties. The
main results could be summarized as follows: (i) sintering at
elevated temperatures from 700 to 900 ꢀC improved the crys-
tallinity and layered structure of the samples, further enhancing
the electrochemical performance. Excessive sintering at a much
higher temperature of 1000 ꢀC led to oversize particles and poor
electrochemical performance. 900 ꢀC was determined to be the
optimal sintering temperature. (ii) The optimum period of
sintering time was determined to be 15 h. Sample H15 delivered
the largest capacity, best rate properties, and relatively high
cycling performance. When varying the period of sintering time
from 5 to 15 h, the impact of the structure and particle size on
the electrochemical performance were ignored, while the
content of Mn3+, electronic conductivity and lithium ion diffu-
sion coefficient in the cycle process are the primary reasons for
the different performances of these samples. In addition, large
particles, a smaller content of Mn3+, relatively low electronic
conductivity and lithium ion diffusion coefficient in the cycle
process are the dominant factors that resulted in the worst
electrochemical performance for the prolonged sintering period
of 20 h.
Fig. 13 (a) Electrochemical impedance spectroscopy (EIS) of different
cells measured when charged to 4.6 V, and (b) profiles of the real parts
of impedance (Zr) vs. uꢁ1/2 from 0.1 to 0.01 Hz and corresponding
linear fitting curves for samples H5, H10, H15, and H20.
The results reported herein provide a new methodology for
obtaining various advanced Li-excess composite materials for
use in high-performance energy storage and conversion devices.
Table 3 Fitted parameters for EIS obtained by Zsimpwin software of
different samples
Acknowledgements
Samples
Re (U)
Rsf (U)
Rct (U)
Rtotal (U)
DLi+(cm2 sꢁ1
)
This work was nancially supported by NSFC (21271171 and
21025104), National Basic Research Program of China
(2011CB935904), and FJPYT (2060303 and 2060203) and FJIRSM
fund (SZD09003-1 and 2010KL002). The authors express their
thanks to Zhenglong Wu from the Analysis and test center of
Beiijng Normal University for XPS measurements.
H5
2.72
6.75
2.60
2.78
10.02
8.69
15.05
6.21
152.10
52.79
57.87
82.49
164.84
68.23
75.52
91.48
2.43 ꢂ 10ꢁ14
6.23 ꢂ 10ꢁ14
1.42 ꢂ 10ꢁ13
5.90 ꢂ 10ꢁ14
H10
H15
H20
two ways: (i) the formation of soluble Mn2+ through the
disproportion reaction of Mn3+ (2Mn3+ / Mn4+ + Mn2+), and (ii)
acid dissolution by HF stemmed from unstable LiPF6 (LiPF6 +
H2O / LiF + 2HF + PF3O).18,60 (2) Phase transition from the
original layered structure to a spinel-like structure. Owing to the
low theoretical capacity, the presence of the spinel structure
would lead to capacity fading. On the other hand, phase tran-
sition might result in the corrosion and fragmentation of the
layered composite cathode,44,61,62 which would reduce the
structural stability. (3) Side reactions at the electrode/electrolyte
interface. Due to the overlap of the d-band of the transition
metals and the oxygen p-bands, the cathode shows oxygen loss
in the form of O2 gas stemmed from the redox reactions
between Co4+ and electrolyte, which may result in the structural
instability of electrode materials.4,22 (4) The precipitation of the
electrolyte decomposition products on the surface of electrode
at high voltage could be another important concern.60
Notes and references
1 J. M. Tarascon and M. Armand, Nature, 2001, 414, 359–367.
2 Y.-G. Guo, J.-S. Hu and L.-J. Wan, Adv. Mater., 2008, 20, 2878–
2887.
3 A. Manthiram, A. V. Murugan, A. Sarkar and T. Muraliganth,
Energy Environ. Sci., 2008, 1, 621–638.
4 J. B. Goodenough and Y. Kim, Chem. Mater., 2009, 22, 587–
603.
5 J. Lu, Q. Peng, W. Wang, C. Nan, L. Li and Y. Li, J. Am. Chem.
Soc., 2013, 135(5), 1649–1652.
6 J. N. Reimers and J. R. Dahn, J. Electrochem. Soc., 1992, 139,
2091–2097.
7 T. Ohzuku and A. Ueda, J. Electrochem. Soc., 1994, 141, 2972–
2977.
8 D. Luo, G. Li, C. Yu, S. l. Yang, J. Zheng, X. Guan and L. Li,
J. Mater. Chem., 2012, 22, 22233–22241.
´
9 K. Kang, Y. S. Meng, J. Breger, C. P. Grey and G. Ceder,
4. Conclusions
Science, 2006, 311, 977–980.
Li1.2Mn0.4Co0.4O2 composites were successfully synthesized by a 10 T. E. Conry, A. Mehta, J. Cabana and M. M. Doeff, Chem.
simple gel-combustion method. This method has many merits.
Mater., 2012, 24, 3307–3317.
1482 | J. Mater. Chem. A, 2014, 2, 1471–1483
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
11 K. C. Jiang, S. Xin, J. S. Lee, J. Kim, X. L. Xiao and Y. G. Guo, 39 X.-J. Guo, Y.-X. Li, M. Zheng, J.-M. Zheng, J. Li, Z.-L. Gong
Phys. Chem. Chem. Phys., 2012, 14, 2934–2939. and Y. Yang, J. Power Sources, 2008, 184, 414–419.
12 Y. Cho, S. Lee, Y. Lee, T. Hong and J. Cho, Adv. Energy Mater., 40 J. M. Zheng, X. B. Wu and Y. Yang, Electrochim. Acta, 2011,
2011, 1, 821–828.
56, 3071–3078.
13 Y. Kim, ACS Appl. Mater. Interfaces, 2012, 4, 2329–2333.
14 Y. Wang, E. Hosono, K. Wang and H. Zhou, Angew. Chem.,
Int. Ed., 2008, 47, 7461–7465.
15 B. Wang, D. Wang, Q. Wang, T. Liu, C. Guo and X. Zhao,
J. Mater. Chem. A, 2013, 1, 135–144.
16 Y. Wang, D. Zhang, X. Yu, R. Cai, Z. Shao, X.-Z. Liao and
Z.-F. Ma, J. Alloys Compd., 2010, 492, 675–680.
17 D. Zhang, R. Cai, Y. Zhou, Z. Shao, X.-Z. Liao and Z.-F. Ma,
Electrochim. Acta, 2010, 55, 2653–2661.
18 M. Hirayama, H. Ido, K. Kim, W. Cho, K. Tamura, J. i. Mizuki
and R. Kanno, J. Am. Chem. Soc., 2010, 132, 15268–15276.
19 J.-Y. Luo, H.-M. Xiong and Y.-Y. Xia, J. Phys. Chem. C, 2008,
112, 12051–12057.
41 R. Santhanam, P. Jones, A. Sumana and B. Rambabu,
J. Power Sources, 2010, 195, 7391–7396.
42 J. Wang, B. Qiu, H. Cao, Y. Xia and Z. Liu, J. Power Sources,
2012, 218, 128–133.
43 K. M. Shaju, G. V. Subba Rao and B. V. R. Chowdari,
Electrochim. Acta, 2002, 48, 145–151.
44 M. Gu, I. Belharouak, J. Zheng, H. Wu, J. Xiao, A. Genc,
K. Amine, S. Thevuthasan, D. R. Baer, J.-G. Zhang,
N. D. Browning, J. Liu and C. Wang, ACS Nano, 2012, 7(1),
760–767.
´
45 A. R. Armstrong, M. Holzapfel, P. Novak, C. S. Johnson,
S.-H. Kang, M. M. Thackeray and P. G. Bruce, J. Am. Chem.
Soc., 2006, 128, 8694–8698.
20 M. M. Thackeray, S.-H. Kang, C. S. Johnson, J. T. Vaughey, 46 Z. Lu and J. R. Dahn, J. Electrochem. Soc., 2002, 149, A815–
R. Benedek and S. A. Hackney, J. Mater. Chem., 2007, 17,
3112–3125.
21 J. Liu, L. Chen, M. Hou, F. Wang, R. Che and Y. Xia, J. Mater.
Chem., 2012, 22, 25380–25387.
A822.
´ ´
47 H. Koga, L. Croguennec, P. Mannessiez, M. Menetrier,
F. Weill, L. Bourgeois, M. Duttine, E. Suard and C. Delmas,
J. Phys. Chem. C, 2012, 116, 13497–13506.
22 N. Yabuuchi, K. Yoshii, S.-T. Myung, I. Nakai and S. Komaba, 48 Z. Li, Y. Wang, X. Bie, K. Zhu, C. Wang, G. Chen and Y. Wei,
J. Am. Chem. Soc., 2011, 133, 4404–4419. Electrochem. Commun., 2011, 13, 1016–1019.
23 G. Singh, R. Thomas, A. Kumar and R. S. Katiyar, 49 S. Kim, C. Kim, Y.-I. Jhon, J.-K. Noh, S. H. Vemuri, R. Smith,
J. Electrochem. Soc., 2012, 159, A410.
24 J.-H. Park, J. Lim, J. Yoon, K.-S. Park, J. Gim, J. Song, H. Park,
K. Y. Chung, M. S. Jhon and B.-W. Cho, J. Mater. Chem., 2012,
22, 25418–25426.
D. Im, M. Park, D. Ahn, Y. Paik and J. Kim, Dalton Trans., 50 W. He, J. Qian, Y. Cao, X. Ai and H. Yang, RSC Adv., 2012, 2,
2012, 41, 3053–3059. 3423–3429.
25 Y. K. Sun, M. G. Kim, S. H. Kang and K. Amine, J. Mater. 51 J. Bareno, M. Balasubramanian, S. H. Kang, J. G. Wen,
˜
Chem., 2003, 13, 319–322.
26 C.-C. Wang and A. Manthiram, J. Mater. Chem. A, 2013, 1,
10209–10217.
27 J. Liu and A. Manthiram, J. Mater. Chem., 2010, 20, 3961–
3967.
28 Q. Y. Wang, J. Liu, A. V. Murugan and A. Manthiram,
J. Mater. Chem., 2009, 19, 4965–4972.
29 H. Li and H. Zhou, Chem. Commun., 2012, 48, 1201–1217.
30 C. J. Jaa, K. I. Ozoemena, M. K. Mathe and W. D. Roos,
Electrochim. Acta, 2012, 85, 411–422.
31 D. Luo, G. Li, X. Guan, C. Yu, J. Zheng, X. Zhang and L. Li,
J. Mater. Chem. A, 2013, 1, 1220–1227.
C. H. Lei, S. V. Pol, I. Petrov and D. P. Abraham, Chem.
Mater., 2011, 23, 2039–2050.
52 A. W. Moses, H. G. G. Flores, J.-G. Kim and M. A. Langell,
Appl. Surf. Sci., 2007, 253, 4782–4791.
53 F. Munakata, H. Takahashi, Y. Akimune, Y. Shichi,
M. Tanimura, Y. Inoue, R. Itti and Y. Koyama, Phys. Rev. B:
Condens. Matter Mater. Phys., 1997, 56, 979–982.
54 C. Yu, X. Guan, G. Li, J. Zheng and L. Li, Scr. Mater., 2012, 66,
300–303.
55 S. Ivanova, E. Zhecheva, R. Stoyanova, D. Nihtianova,
S. Wegner, P. Tzvetkova and S. Simova, J. Phys. Chem. C,
2011, 115, 25170–25182.
32 C. Yu, G. Li, X. Guan, J. Zheng and L. Li, Electrochim. Acta, 56 S. Ardizzone, C. L. Bianchi and D. Tirelli, Colloids Surf., A,
2012, 61, 216–224.
1998, 134, 305–312.
33 B. H. Toby, J. Appl. Crystallogr., 2001, 34, 210–213.
34 X. Zhang, D. Luo, G. Li, J. Zheng, C. Yu, X. Guan, C. Fu,
X. Huang and L. Li, J. Mater. Chem. A, 2013, 1, 9721–9729.
35 M. Park, X. Zhang, M. Chung, G. B. Less and A. M. Sastry,
J. Power Sources, 2010, 195, 7904–7929.
57 Q.-H. Wu, M. Liu and W. Jaegermann, Mater. Lett., 2005, 59,
1980–1983.
58 D. Andre, M. Meiler, K. Steiner, C. Wimmer, T. Soczka-Guth
and D. U. Sauer, J. Power Sources, 2011, 196, 5334–5341.
59 G. G. Amatucci, J. M. Tarascon and L. C. Klein, Solid State
Ionics, 1996, 83, 167–173.
36 A. Abouimrane, O. C. Compton, H. Deng, I. Belharouak,
D. A. Dikin, S. T. Nguyen and K. Amine, Electrochem. Solid- 60 K. T. Lee, S. Jeong and J. Cho, Acc. Chem. Res., 2013, 46(5),
State Lett., 2011, 14, A126. 1161–1170.
37 X. Zhang, C. Yu, X. Huang, J. Zheng, X. Guan, D. Luo and 61 J. Zheng, M. Gu, J. Xiao, P. Zuo, C. Wang and J.-G. Zhang,
L. Li, Electrochim. Acta, 2012, 81, 233–238. Nano Lett., 2013, 13(8), 3824–3830.
38 C. Yu, H. Wang, X. Guan, J. Zheng and L. Li, J. Alloys Compd., 62 B. Song, Z. Liu, M. O. Lai and L. Lu, Phys. Chem. Chem. Phys.,
2013, 546, 239–245.
2012, 14, 12875–12883.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 1471–1483 | 1483