Li4/3Ni1/3Mo1/3O2 – LiNi1/2Mn1/2O2 binary system as high capacity positive electrode materials for rechargeable lithium batteries

Article abstract of DOI:10.1149/2.0661807jes

A binary system of x Li_4/3Ni_1/3Mo_1/3O₂ – (1–x) LiNi_1/2Mn_1/2O₂ is studied as high-capacity positive electrode materials for rechargeable lithium batteries. Structural and electrochemical properties of oxides with different compositions in this binary system are examined. Mo ordering is retained for 1 ≤ x ≤ 1/3 with a monoclinic symmetry and disappears for x ≤ 1/6 with a rhombohedral symmetry. Compared with Li_4/3Ni_1/3Mo_1/3O₂, partial substitution of Mn for Mo lead to the improvement of reversible capacity and reduction of polarization. For Li_6/5Ni_2/5Mn_1/5Mo_1/5O₂ (x = 1/3) and Li_9/8Ni_7/16Mn_5/16Mo_1/8O₂ (x = 1/6), high reversible capacities of around 200 mAh g^?1 are obtained. Improved cycling performance is achieved through the optimization of voltage ranges. Further structural characterization by ex–situ XRD reveals that the improved reversibility for the Mn-substituted samples mainly results from the suppression of Mo migration during cycling, probably associated with partial oxygen loss.

Full text of DOI:10.1149/2.0661807jes

Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
A1357
Li4/3Ni1/3Mo1/3O2 – LiNi1/2Mn1/2O2 Binary System as High
Capacity Positive Electrode Materials for Rechargeable Lithium
Batteries
∗,z
Wenwen Zhao, Kazuma Yamaguchi, Takahito Sato, and Naoaki Yabuuchi
Department of Applied Chemistry, Tokyo Denki University, Adachi, Tokyo 120-8551, Japan
A binary system of x Li4/3Ni1/3Mo1/3O2 – (1–x) LiNi1/2Mn1/2O2 is studied as high-capacity positive electrode materials for
rechargeable lithium batteries. Structural and electrochemical properties of oxides with different compositions in this binary system
are examined. Mo ordering is retained for 1 ≤ x ≤ 1/3 with a monoclinic symmetry and disappears for x ≤ 1/6 with a rhombohedral
symmetry. Compared with Li4/3Ni1/3Mo1/3O2, partial substitution of Mn for Mo lead to the improvement of reversible capacity and
reduction of polarization. For Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) and Li9/8Ni7/16Mn5/16Mo1/8O2 (x = 1/6), high reversible capacities
of around 200 mAh g−1 are obtained. Improved cycling performance is achieved through the optimization of voltage ranges. Further
structural characterization by ex–situ XRD reveals that the improved reversibility for the Mn-substituted samples mainly results from
the suppression of Mo migration during cycling, probably associated with partial oxygen loss.
© The Author(s) 2018. Published by ECS. This is an open access article distributed under the terms of the Creative Commons
Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any
medium, provided the original work is properly cited. [DOI: 10.1149/2.0661807jes]
Manuscript submitted March 6, 2018; revised manuscript received April 13, 2018. Published May 5, 2018.
Lithium-ion batteries (LIBs) have dominated the market of
portable electronic devices during the past two decades. As the de-
velopment in the technology of material science, LIBs are applied
to the electric vehicles (EV), and its market is being enlarged in a
past few years. To enable the long range driving for EVs, the fur-
ther development of rechargeable batteries with high energy density
is crucially important. Searching and developing positive electrode
materials with high energy density are of great importance to enlarge
the overall energy density of LIBs. As a traditional group of positive
electrode materials for LIBs, lithium transition metal oxides LiMeO2
(Me = Mn, Co, and Ni) have been intensively studied.1–12 Among
substitution by other electrochemically active transition metal ions
is expected to be an effective approach. Because of the structural
compatibility between Li4/3Ni1/3Mo1/3O2 and LiNi1/2Mn1/2O2 pos-
sessing the same oxygen packing, formation of solid solution of
two materials is expected. Therefore, in this study, a binary system
of Li4/3Ni1/3Mo1/3O2–LiNi1/2Mn1/2O2 is targeted and synthesized as
potential high capacity positive electrode materials for LIBs. Impact
of the substitution of Mn for Mo on the electrode performance and
structural stability is also systematically studied. From these results,
the strategy to design high capacity Li-excess oxides for advanced
LIBs is further discussed.
them, LiMnO2,1–3 LiCoO2,9,10 LiNiO2,13,14 LiNi1/2Mn1/2O2,7,15 and
6,12,16,17
LiNi1/3Co1/3Mn1/3O2
have drawn much attention from both
Experimental
industrial and academic aspects. However, further increase in the re-
versible capacities for these electrode materials on the basis of the
cationic redox reaction is restricted in these layered transition metal
oxides.
Material synthesis.—The target materials were synthesized via
solid-state calcination method. Starting materials consisting of
Li2CO3 (98.5%, Kanto Kagaku), Ni(CO3)2•Ni(OH)2•4H2O (Kishida
Chemical), MnCO3 (Kishida Chemical) and Li2MoO4 were mixed
and ground by using a mortar and pestle. Note that MoO3 easily
reacts with lithium salts and forms compounds with low melting tem-
perature (∼500◦C),30 and Li2MoO4 with high melting temperature
(700◦C) is, therefore, suitable for the precursor. Li2MoO4 was synthe-
sized by heating of a mixture of MoO3 (99.5%, Kanto Kagaku) and
LiOH (95.0%, Wako). The obtained mixtures were wet ball-milled
with methanol at 300 rpm for 5 h to enable uniform mixing. Thus
obtained ball-milled precursors after drying were pressed into pellets
followed by calcination at 1050◦C for 5 h in air.
As another group of positive electrode materials, Li-excess man-
ganese oxides, Li2MnO3 and its derivatives, have been extensively
studied as high capacity positive electrode materials for LIBs.18–21
The successful utilization of both cationic and anionic redox reaction
enables the development of high-capacity electrode materials via the
enrichment of Li ions in the structures.22 This conception has also been
proposed and experimentally verified with Li3NbO4 and Li4MoO5 as
Li-excess host structures for designing high-capacity positive elec-
trode materials.23,24 Similar to Li2MnO3, both Li3NbO4 and Li4MoO5
can be classified as cation-ordered rocksalt–type structures. However,
pure Li3NbO4 and Li4MoO5 are electrochemically inactive because
of the low electronic conductivity of Nb and Mo due to the absence
of d-electrons in a conduction band. To enhance the electrochemical
activity of these materials, partial substitution of Nb and Li by 3d
transition metals (Ni2+, Co2+, V3+, Fe3+ and Mn3+) has been system-
ically studied, and it has been demonstrated that promising electrode
performance can be achieved through the proper selection of substi-
tuted metal ions.25–27 The same approach has been also applied to
Li4MoO5 system, such as Li4MoO5–LiFeO2,28 and Li4MoO5–NiO29
and Li-Ni-Ti-Mo-O23 are also examined as electrode materials.
In the binary system of Li4MoO5–NiO, Li4NiMoO6
(Li4/3Ni1/3Mo1/3O2) with a space group of C2/m delivers an
initial charge capacity of over 300 mAh g−1. However, large
polarization and irreversible phase transition, associated with oxygen
loss, have been observed for Li4/3Ni1/3Mo1/3O2 by charge to high
voltage.29 To improve the electrode performance of Li4/3Ni1/3Mo1/3O2,
Characterization methods.—Crystal structures of the samples
were examined by using an X-ray diffractometer (XRD) equipped
with a high-speed one dimensional detector (D2 Phaser, Bruker).
Non-monochromatized Cu Kα radiation was utilized as an X-ray
source with a nickel filter. Synchrotron XRD data were collected at
the beamline BL02B2 at SPring-8 synchrotron facility in Japan.31 The
measurement was conducted using an automatic powder diffraction
system for Debye-Scherrer geometry using a sample in glass cap-
illary. The wavelength used is 0.5003 Å, which was calibrated with
CeO2 as a reference sample. Rietveld refinement was carried out using
RIETAN-FP software.32 The morphological features of the samples
were observed using a scanning electron microscope (SEM, JEOM,
JCM – 6000).
Electrochemical characterization.—Electrode performance of the
samples was examined in two-electrode type Li cells (Tomcell Japan,
type TJ-AC). The composite positive electrodes consisted of 80 wt%
as-prepared oxide powder, 10 wt% acethylene black (AB; Denka
∗
Electrochemical Society Member.
zE-mail: yabuuchi@mail.dendai.ac.jp
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1358
Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
Figure 1. XRD patterns of x Li4/3Ni1/3Mo1/3O2 – (1 – x)
LiNi1/2Mn1/2O2 binary phase. Schematic illustrations of
crystal structures are drawn by using VESTA.40
HS-100) and 10 wt% poly(vinylidene fluoride) (Kureha #1100) with
N-methylpyrrolidone pasted on aluminum foil used as a current
collector. The electrodes were dried at 120◦C in vacuum. Metallic
lithium (Honjo Metal) was used as a negative electrode. Electrolyte
solution used was 1.0 mol dm−3 LiPF6 dissolved in ethylene car-
bonate:dimethyl carbonate (1:1 by volume) (Battery grade, Kishida
Chemical). A polyolefin microporous membrane was used as separa-
tor. Electrochemical cells were assembled in an Ar-filled glove box
and cycled at a rate of 10 mA g−1 at room temperature. The sam-
ples for ex–situ XRD study were prepared by removing composite
electrodes from the cells after washing with battery grade dimethyl
carbonate (DMC). All procedures were carried out in the Ar-filled
glove box.
the β angle continuously decreases with Mn substitution. As shown
in the enlarged XRD patterns (Supporting Figure S1), Mo-rich sam-
¯
ples clearly show prominent peak splitting of 202 and 131 diffraction
lines. Since bhex./a hex. ratios for the all samples in Figure 2 are close to
√
3 for an ideal triangle lattice, in-plane distortion is not responsible
for the peak splitting. Instead, the increase in the β angle results in
¯
the peak splitting of the (202) and (131) diffraction lines. Supporting
110.5
110.0
109.5
Results and Discussion
β
109.0
XRD patterns of x Li4/3Ni1/3Mo1/3O2 – (1–x) LiNi1/2Mn1/2O2 bi-
nary system with different chemical compositions are shown in Figure
1. LiNi1/2Mn1/2O2 (x = 0) crystallizes into an α-NaFeO2-type layered
4.85
4.80
4.75
¯
structure with a space group of R3 m. Nearly 10% of an anti-site defect
is noted between 3a and 3b octahedral sites.33 Another end-member,
Li4/3Ni1/3Mo1/3O2 (x = 1), which is a 1:1 mixture of Li4MoO5 and
NiO, has an analogue structure with Li5ReO6. The detailed crystal
structure of Li4MoO5 is found in our previous work.29 It is found that
both two end-members have similar lattice parameters reported in the
literature.29,33 Substitution of Mn for Mo in Li4/3Ni1/3Mo1/3O2 does
not significantly change the host structure. Further increase in the Mn
content leads to the decrease of the intensity of superlattice diffraction
peaks, related to the Mo ordering, located between 2θ = 20–35◦.
Moreover, from the enlarged XRD patterns (Supporting Figure S1)
between 2θ = 41– 45◦, it is noticeable that diffraction lines at 2θ = 44◦
merge into a single peak as the Mn content increases. Such continuous
peak evolution indicates increased symmetry (from a monoclinic to
rhombohedral lattice) for the samples as the Mn content increases.
Variation of lattice parameters of the samples with different Mn con-
tents is summarized in Figure 2. The calculated metal-metal (Me-Me)
interatomic distances and inter-slab distances are presented. From
x = 1.0 to 0.5, almost no change is observed in metal-metal inter-
atomic distances whereas interslab distances slightly increase from
4.76 to 4.78 Å. This trend would originate from the loss of the three-
dimensional structure and gradual formation of the layered structure.
Therefore, this binary system cannot obey the Vegard’s law as an
ideal solid solution phase and shows complicated phase evolution.
Further decrease in the x value results in the contraction of the lat-
tice for both metal-metal and interslab distances as expected. Note
that the monoclinic symmetry is retained for x = ∼0.2. Furthermore,
4.70
1.76
√3
1.74
1.72
1.70
3.00
2.95
2.90
2.85
1.0
0.8
0.6
0.4
0.2
0.0
x in
x Li4/3Ni1/3Mo1/3O2 - (1-x) LiNi1/2Mn1/2O2
Figure 2. Evolution of crystallographic parameters for x Li4/3Ni1/3Mo1/3O2
– (1 – x) LiNi1/2Mn1/2O2 binary system; metal-metal distances, inter-slab
distance and β angle as a function of x values.
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
A1359
Observed
Calculated
Difference
Li1.2Ni0.4Mn0.2Mo0.2O2
Li2MoO4 (5%)
5
10
15
20
25
30
35
40
2θ / deg. (
λ
= 0.5 Å)
Figure 3. The result of Rietveld refinement for Li6/5Ni2/5Mn1/5Mo1/5O2 (x =
1/3). Refined structural parameters are summarized in Table I.
Figure S2 shows simulated XRD patterns of Li4/3Ni1/3Mo1/3O2 with
different β angles and the fixed b/a ratio. It is noticeable that merg-
¯
ing (202) and (131) lines originates from the decrease in β values.
This observation is consistent with the fact that the samples con-
tain non-Jahn-Teller ions in the structure. The calculated β angle
for Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) is 109.53◦, which is slightly
larger than that of LiNi1/2Mn1/2O2 (β = 109.23◦) calculated from the
rhombohedral lattice. Such lattice distortion along c-axis direction is
continuously relaxed by Mn substitution for Mo ions, and completely
lost for LiNi0.5Mn0.5O2 as shown in Figure 2.
Figure 3 shows a result of Rietveld refinement for a synchrotron
XRD pattern of Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) by using RIETAN-
FP. Li2MoO4 as a minor phase (approximately 5% by mass) is also
found. Detailed structural parameters and refined results are summa-
rized in Table I. A good refinement result is obtained using a space
group of C2/m. From the refined results, it is found that Li and Ni
ions are randomly distributed in different octahedral sites (4 g, 4 h
and 2d sites) due to the similar ionic radii of Li and Ni2+ ions (Li+;
0.90 Å and Ni2+; 0.83 Å).34 The refined result also indicates that both
Mo and Mn ions tend to preferentially occupy 4g site, and ordering
Figure 4. SEM images of x Li4/3Ni1/3Mo1/3O2 – (1 – x) LiNi1/2Mn1/2O2
binary phase and elemental mappings by EDX for Li6/5Ni2/5Mn1/5Mo1/5O2 (x
= 1/3). Note that magnification of the sample (x = 1) is different from other
samples.
between Mo/Mn ions, which are smaller ions (Mn4+; 0.67 Å and
Mo6+; 0.73 Å)34 is noted. In addition, disordering of Li/Ni ions in 4 h
and 2d sites is also clearly evidenced, similar to Li4/3Ni1/3Mo1/3O2.
A partial distribution of Mo in 2a site is also noted from the refined
result. Therefore, the difference in atomic scattering factors for 2a and
4g sites results in the appearance of superlattice lines based on the
layered structure.
Table I. Refined structural parameters of Li6/5Ni2/5Mn1/5Mo1/5O2
(x = 1/3) by the Rietveld method.
Atom
Site
Occupancy
x
y
z
B† / Å2
Particle morphology of the samples was observed by SEM.
Figure 4 displays SEM images of the samples and EDX mapping of
Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3). It is noticeable that the samples with
the high Mo content have round–shaped morphology with an average
particle size of > 2 μm. As the Mn content increases, the particle size
is drastically reduced and changes to uniform size for LiNi1/2Mn1/2O2
(x = 0). Moreover, EDX mappings of Li6/5Ni2/5Mn1/5Mo1/5O2 (x =
1/3) suggest that Ni, Mn and Mo are essentially uniformly distributed,
indicating formation of solid solution between Li4/3Ni1/3Mo1/3O2 and
LiNi1/2Mn1/2O2. Nevertheless, the presence of Mo-rich regions is
noted, which corresponds to the impurity found by Rietveld analy-
sis as shown in Figure 3. This fact suggests that lithium vaporization
during calcination and the formation of the lithium deficient phase.
Similarly, a uniform distribution of Ni, Mn and Mo is also found in
Li9/8Ni7/16Mn5/16Mo1/8O2 (x = 1/6) (Supporting Figure S3).
Li
Ni
Mn
Mo
Li
Ni
Li
Ni
Li
4g
4g
4g
4g
2d
2d
4h
4h
2a
2a
4i
0.116
0.284
0.30
0
0
0
0
0
0
0
0
0
0.671
0.671
0.671
0.671
0.5
0
0
0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.9
0.9
0.30
0
0.815
0.185
0.911
0.089
0.728
0.272
1.0
0.5
0.5
0.5
0.5
0
0.5
0.163
0.163
0
0
0
Ni
O
O
0
0
0.243
0.250
0.795
0.793
8j
1.0
0.335
Monoclinic, space group C2/m, a = 5.0639(3) Å, b = 8.732(1) Å, and
c = 5.0685(3) Å, β = 109.49 (1)◦ and V = 211.35 Å3.
Rwp = 5.75% and RB = 4.19%.
To examine the electrode performance of the as-prepared sam-
ples, galvanostatic charge/discharge cycle tests were performed in Li
†Not refined.
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1360
Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
5.0
1st
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
2nd
x = 1
1st
2nd
x = 1/2
1st
Figure 5. Charge/discharge profiles of x Li4/3Ni1/3Mo1/3O2 – (1 – x)
LiNi1/2Mn1/2O2 binary phase.
2nd
x = 1/3
x = 1/6
1st
2nd
1st
2nd
50
x = 0
0
100
150
200
250
300
350
Capacity / mAh g-1
Ni2+/Ni4+-based
228 mAh g-1
Li contents-based
342 mAh g-1
cells. Figure 5 shows the initial charge/discharge profiles of all sam-
ples shown in Figure 1. The voltage profiles of two end-members,
Li4/3Ni1/3Mo1/3O2 and LiNi1/2Mn1/2O2, are identical to the reported
data.29,33 LiNi1/2Mn1/2O2 shows quite small polarization with a re-
versible capacity of 200 mAh g−1. However, an initial charge capacity
is less than 225 mAh g−1 even with cutoff voltage of 4.8 V, which
is due to limited charge capacity based on Ni2+/Ni4+ cationic redox
reaction. For Li4/3Ni1/3Mo1/3O2, enrichment of the Li content in the
framework structure increases the initial charge capacity to around 330
mAh g−1, which is much larger than that of LiNi1/2Mn1/2O2. The ob-
served charge capacity reaches 86% of the theoretical capacity based
on the Li content (385 mAh g−1). The observed capacity exceeds the
theoretical capacity based on Ni2+/Ni4+ cationic redox reaction (192
mAh g−1), which reveals that redox reaction of oxide ions is involved
for the large charge capacity.25 Nevertheless, a large irreversible ca-
pacity together with huge polarization is also noticeable. Similar to
Li2MnO3 and other Li-excess based positive electrode materials, this
observation can be explained by the irreversible partial oxygen loss
during the initial charge process, leading to degradation of electrode
performance. Among the tested samples, the sample of x = 1/6 shows
the largest initial reversible capacity of 240 mAh g−1 with relatively
small polarization and good capacity retention. The charge/discharge
profiles of samples from 2nd to 10th cycle are shown in Supporting
Figure S4. It is noted that a voltage plateau at approximately 2 V is
observed for the Mo-containing samples. A similar voltage plateau
is also observed for the Li4MoO5–LiFeO2 binary system.28 The ob-
served large polarization and the origin of the 2 V plateau will be
discussed in the later section.
The influence of cutoff voltage ranges on the reversible capac-
ity and cycle performance was further systemically examined. Initial
charge/discharge curves of Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) with
different cutoff voltage is shown in Figure 6. At the higher cutoff
voltage, high reversible capacity is obtained due to the activation
of oxide ion redox. On the other hand, an increased irreversible
capacity and enlarged polarization are also noted as mentioned be-
fore. Similar behavior is also found for Li9/7Ni5/14Mn1/14Mo2/7O2
(x = 2/3) and Li5/4Ni3/8Mn1/8Mo1/4O2 (x = 1/2) samples (Support-
ing Figure S5). The dependence of the reversible capacity upon
cycling on the different cutoff voltage ranges is well reflected
in Figure 6b. Capacity fading is observed at the initial 10 cy-
cles when charged above 4.4 V. In contrast, excellent cyclability
(a)
6.0
5.0
4.8 V
4.4 V
4.11 V
4.0
3.0
2.0
1.0
0.0
3.8 V
1st cycle
0
100
200
300
Capacity / mAh g-1
250
200
150
100
50
(b)
4.8 V charge
4.4 V charge
3.8 V charge
4.11 V charge
0
0
5
10
15
20
25
30
Cycle Number
Figure 6. Initial charge/discharge profiles of Li6/5Ni2/5Mn1/5Mo1/5O2 at dif-
ferent voltage ranges (a), and capacity retention is also shown in (b). Dotted
lines correspond to theoretical capacities.
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
A1361
Al
1.5 V discharge
(after charge to 4.8 V)
4.8 V charge
1.5 V discharge
(after charge to 4.1 V)
Figure 7. Changes in XRD patterns of Li4/3Ni1/3Mo1/3O2 (x = 1/3)
on the electrochemical cycle with different cutoff voltages.
4.1 V charge
Pristine
10
20
30
40
50
60
70
2θ / deg. (Cu Kα)
4.8 V in Figure 5 suggests that partial reduction of Mn35 and Mo28
after the irreversible oxygen loss. On the other hand, the partially Mn-
substituted sample, Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) shows different
behavior when compared with Li4/3Ni1/3Mo1/3O2 during charge. The
intensity of the 110 superlattice line is clearly visible even after charge
to 4.8 V even though the peak intensity is relatively weakened. This
fact indicates that oxygen loss is partly suppressed compared with
the Mo-rich sample (x = 1). Moreover, structural change is further
suppressed by lowering cutoff voltage to 4.1 V as shown in Figure
7. The results are also consistent with the electrochemical data in
Figure 6. Indeed, the voltage plateau at 2 V is much shorter for the
Mn-substituted sample (x = 1/3). Comparison of the ex–situ XRD
results for Li4/3Ni1/3Mo1/3O2 and Li6/5Ni2/5Mn1/5Mo1/5O2 enables us
to conclude that irreversible structural change associated with oxygen
loss at the initial cycle is partially suppressed through Mn substitution,
leading to better electrode performance. Note that Li contents in the
structure in both samples are different, and this fact also influences
the structural stability on charge. This effect will be further discussed
elsewhere.
is achieved at the cutoff voltage of 4.1 V. Such dependency of
the cutoff voltage on cyclability is further discussed in the next
section.
To examine the origin of large polarization and insufficient
cyclability as electrode materials for the Mo-substituted samples,
ex–situ XRD and XAS spectroscopy were utilized. Ex-situ XRD
patterns of Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3) after electrochemical
cycle with different cutoff voltage are shown in Figure 7. For charged
Li4/3Ni1/3Mo1/3O2 (x = 1), the diffraction peaks shift to high angle
side, accompanied with the disappearance of the superlattice lines
originating from the Mo ordering as shown in Supporting Figure S6.
Additionally, upon discharging, intensity of 20–2 and 131 diffraction
lines is intensified, suggesting the progress on the cation disordering in
the ccp oxygen lattice. Loss of the cation ordering is also observed for
the Li2MnO3-based electrode materials, and this process is triggered
by oxygen loss and structural reconstruction process.35 Therefore,
similar oxygen loss on charge to 4.8 V is anticipated in this system,
and such trend is clearly visible from charge/discharge profiles. The
long voltage plateau observed on discharge at 2 V after charge to
Li4/3Ni1/3Mo1/3O2
LiNi1/2Mn1/2O2
Reducꢀve Coupling
and Oxygen Loss
Highly Reversible
Ef
Ef
eg
Ef
Ef
eg
eg
eg
Figure 8. Proposed charge compensation mechanisms for
LiNi1/2Mn1/2O2 and Li4/3Ni1/3Mo1/3O2.
2p
2p
t2g
t2g
t2g
t2g
Ni3+
On-
Ni2+
O2-
Ni4+
O2-
Ni2+
O2-
Before
Li extracꢀon
Aꢁer
Li extracꢀon
Before
Li extracꢀon
Aꢁer
Li extracꢀon
Caꢀonic redox
Caꢀonic and Anionic Redox
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1362
Journal of The Electrochemical Society, 165 (7) A1357-A1362 (2018)
Based on the experimental results obtained from this study
ORCID
and our previously reported studies on Li-excess cation-disordered
rocksalt-type materials, charge compensation mechanisms for
Li4/3Ni1/3Mo1/3O2–LiNi1/2Mn1/2O2 binary system are summarized in
Figure 8. For the conventional stoichiometric layered materials, like
LiNi1/2Mn1/2O2 and LiNiO2, the Fermi level of the samples mainly
consists of Ni eg band with partial contribution of oxygen 2p band.
Therefore, highly reversible cationic redox reaction is evidenced by
soft X-ray absorption spectroscopy.14 Contribution of oxygen for the
redox process is limited even though high covalency for Ni4+ results
in the charge redistribution between metal and oxygen bonds. As a
result, lithium extraction from the structure leads to partial oxidation
of oxygen, but this process is essentially classified as cationic redox.
On the other hand, for another end-member, Li4/3Ni1/3Mo1/3O2, the
reaction mechanism is different from that of the cationic redox reac-
tion. Enrichment of Li and Mo ions, which have a high ionic character
with a less covalent character, lowers the energy of oxygen 2p band.22
This process is also explained by the formation of unhybridized Li-
O-Li states, which have a non-bonding character.36,37 Since energy of
oxygen 2p and nickel eg bands is very close for each other, competition
of anionic and cationic redox is unavoidable. Nevertheless, oxidation
of the oxygen 2p band, namely creation of holes in the oxygen 2p
band,38 cannot be energetically stabilized for highly covalent system,
like Ni4+, and results in the reductive coupling.25,39 Similar phenom-
ena are observed for Li1.3Nb0.3Fe0.4O2 and Li1.2Ti0.4Fe0.4O2, in which
the partially filled eg bands are created coupled with ions possessing
Naoaki Yabuuchi https://orcid.org/0000-0002-9404-5693
References
1. A. R. Armstrong and P. G. Bruce, Nature, 381, 499 (1996).
2. P. G. Bruce, A. R. Armstrong, and R. L. Gitzendanner, Journal of Materials Chem-
istry, 9, 193 (1999).
3. L. Croguennec, P. Deniard, R. Brec, and A. Lecerf, Journal of Materials Chemistry,
7, 511 (1997).
4. Y. M. Choi, S. I. Pyun, J. S. Bae, and S. I. Moon, Journal of Power Sources, 56, 25
(1995).
5. K. Nakamura, H. Ohno, K. Okamura, Y. Michihiro, I. Nakabayashi, and T. Kanashiro,
Solid State Ionics, 135, 143 (2000).
6. T. Ohzuku and Y. Makimura, Chemistry Letters, 30, 642 (2001).
7. T. Ohzuku and Y. Makimura, Chem Lett, 30, 744 (2001).
8. T. Ohzuku, A. Ueda, M. Nagayama, Y. Iwakoshi, and H. Komori, Electrochimica
Acta, 38, 1159 (1993).
9. E. I. Santiago, A. V. C. Andrade, C. O. Paiva-Santos, and L. O. S. Bulho˜es, Solid
State Ionics, 158, 91 (2003).
10. Y. Shao-Horn, L. Croguennec, C. Delmas, E. C. Nelson, and M. A. O’Keefe, Atomic
resolution of lithium ions in LiCoO2, in Nature Materials, 464 (2003).
11. G. Vitins, Journal of The Electrochemical Society, 144, 2587 (1997).
12. N. Yabuuchi and T. Ohzuku, Novel lithium insertion material of LiCo1/3Ni1/3Mn1/3O2
for advanced lithium-ion batteries, in Journal of Power Sources, 171
(2003).
13. T. Ohzuku, A. Ueda, and M. Nagayama, Journal of The Electrochemical Society,
140, 1862 (1993).
14. W. S. Yoon, K. Y. Chung, J. McBreen, D. A. Fischer, and X. Q. Yang, Journal of
Power Sources, 163, 234 (2006).
15. Y. Makimura and T. Ohzuku, Lithium insertion material of LiNi1/2Mn1/2O2
for advanced lithium-ion batteries, in Journal of Power Sources, 156
(2003).
16. S. Kabi and A. Ghosh, Mater. Res. Bull., 48, 3405 (2013).
17. S. Kabi and A. Ghosh, The European Physical Journal B, 79, 377 (2011).
18. D. Y. W. Yu and K. Yanagida, Journal of The Electrochemical Society, 158, A1015
(2011).
19. D. Y. W. Yu, K. Yanagida, Y. Kato, and H. Nakamura, J. Electrochem. Soc., 156,
A417 (2009).
20. Z. H. Lu, D. D. MacNeil, and J. R. Dahn, Electrochemical and Solid State Letters, 4,
A191 (2001).
a high covalent nature, i.e., Fe4+ 25
Further oxidation of oxide ions
.
results in the oxygen loss as the irreversible process. Therefore, less
covalent ions, like Mn4+, are needed to stabilize the reversible solid-
state redox reaction of oxide ions. Further detailed and systematic
studies on the evolution of electronic structures of transition met-
als and oxygen upon electrochemical cycles, including irreversible
oxygen loss, will open a path to design high performance positive
electrode materials for rechargeable lithium battery applications in
the future.
21. Z. H. Lu and J. R. Dahn, J. Electrochem. Soc., 149, A815 (2002).
22. N. Yabuuchi, Chemistry Letters, 46, 412 (2017).
23. J. Lee, D.-H. Seo, M. Balasubramanian, N. Twu, X. Li, and G. Ceder, Energy Environ.
Conclusions
Sci., 8, 3255 (2015).
24. N. Yabuuchi, M. Takeuchi, M. Nakayama, H. Shiiba, M. Ogawa, K. Nakayama,
T. Ohta, D. Endo, T. Ozaki, T. Inamasu, K. Sato, and S. Komaba, Proceedings of the
National Academy of Sciences, 112, 7650 (2015).
25. N. Yabuuchi, M. Nakayama, M. Takeuchi, S. Komaba, Y. Hashimoto, T. Mukai,
H. Shiiba, K. Sato, Y. Kobayashi, A. Nakao, M. Yonemura, K. Yamanaka,
K. Mitsuhara, and T. Ohta, Nature Communications, 7, 13814 (2016).
26. N. Yabuuchi, M. Takeuchi, S. Komaba, S. Ichikawa, T. Ozaki, and T. Inamasu,
Chemical Communications, 52, 2051 (2016).
27. M. Nakajima and N. Yabuuchi, Chem. Mater., 29, 6927 (2017).
28. T. Matsuhara, Y. Tsuchiya, K. Yamanaka, K. Mitsuhara, T. Ohta, and N. Yabuuchi,
Electrochemistry, 84, 797 (2016).
29. N. Yabuuchi, Y. Tahara, S. Komaba, S. Kitada, and Y. Kajiya, Chem. Mater., 28, 416
(2016).
30. M. Moser, D. Klimm, S. Ganschow, A. Kwasniewski, and K. Jacobs, Crystal Research
and Technology, 43, 350 (2008).
31. S. Kawaguchi, M. Takemoto, K. Osaka, E. Nishibori, C. Moriyoshi, Y. Kubota,
Y. Kuroiwa, and K. Sugimoto, Review of Scientific Instruments, 88, 085111
(2017).
32. F. Izumi and K. Momma, Solid State Phenomena, 130, 15 (2007).
33. N. Yabuuchi, S. Kumar, H. H. Li, Y.-T. T. Kim, and Y. Shao-Horn, Journal of The
Electrochemical Society, 154, A566 (2007).
34. R. D. Shannon, Acta Crystallogr., Sect. A, 32, 751 (1976).
35. N. Yabuuchi, K. Yoshii, S.-T. Myung, I. Nakai, and S. Komaba, J. Am. Chem. Soc.,
133, 4404 (2011).
36. D.-H. Seo, J. Lee, A. Urban, R. Malik, S. Kang, and G. Ceder, Nature Chemistry, 8,
692 (2016).
37. M. Okubo and A. Yamada, ACS Appl. Mater. Interfaces, 9, 36463 (2017).
38. K. Luo, M. R. Roberts, R. Hao, N. Guerrini, D. M. Pickup, Y.-S. Liu, K. Edstro¨m,
J. Guo, A. V. Chadwick, L. C. Duda, and P. G. Bruce, Nature Chemistry, 8, 684
(2016).
39. M. Saubanere, E. McCalla, J. M. Tarascon, and M. L. Doublet, Energy & Environ-
mental Science, 9, 984 (2016).
40. K. Momma and F. Izumi, Journal of Applied Crystallography, 44, 1272
(2011).
Synthesis and electrochemical properties of Li4/3Ni1/3Mo1/3O2–
LiNi1/2Mn1/2O2 binary system as high capacity positive electrode
materials for rechargeable LIBs have been reported. Continuous
structural changes are noticed as increase in the Mn content in
this binary system. It has been demonstrated that the electrode per-
formance of Li4/3Ni1/3Mo1/3O2 (x = 1) samples is effectively im-
proved through Mn substitution. Li6/5Ni2/5Mn1/5Mo1/5O2 (x = 1/3)
and Li9/8Ni7/16Mn5/16Mo1/8O2 (x = 1/6) deliver reversible capacities
of approximately 200 mAh g−1. Furthermore, large polarization is
effectively reduced for the Mn-substituted samples. From the electro-
chemical data coupled with ex-situ XRD data, it has been proposed
that the origin of the enhanced reversibility for Mn-substituted sam-
ples originates from the suppression of Mo migration during cycling
associated with irreversible oxygen loss. The suppression of the ir-
reversible oxygen loss is the key to stabilize anionic redox reaction
for Li-excess oxides for high capacity positive electrode materials for
LIBs.
Acknowledgments
The author acknowledges research funding from Advanced Low
Carbon Technology Research and Development Program of the Japan
Science and Technology Agency (JST) Special Priority Research Area
“Next-Generation Rechargeable Battery”. The synchrotron radiation
experiments were performed at the BL02B2 of SPring-8 with the ap-
proval of the Japan Synchrotron Radiation Research Institute (JASRI)
(Proposal 2017B1218).
Downloaded on 2018-05-12 to IP 80.82.77.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).