Article

Cite This: Acc. Chem. Res. 2018, 51, 290−298 pubs.acs.org/accounts

Probing the Complexities of Structural Changes in Layered Oxide

Cathode Materials for Li-Ion Batteries during Fast Charge−Discharge
Cycling and Heating
Published as part of the Accounts of Chemical Research special issue “Energy Storage: Complexities Among
Materials and Interfaces at Multiple Length Scales”.
Enyuan Hu,†,§ Xuelong Wang,†,‡,§ Xiqian Yu,*,‡ and Xiao-Qing Yang*,†
†
Chemistry Division, Brookhaven National Laboratory, Upton, New York 11973, United States
‡
Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190,
China

CONSPECTUS: The rechargeable lithium-ion battery (LIB) is the most

promising energy storage system to power electric vehicles with high
energy density and long cycling life. However, in order to meet customers’
demands for fast charging, the power performances of current LIBs need
to be improved. From the cathode aspect, layer-structured cathode
materials are widely used in today’s market and will continue to play
important roles in the near future. The high rate capability of layered
cathode materials during charging and discharging is critical to the power
performance of the whole cell and the thermal stability is closely related to
the safety issues. Therefore, the in-depth understanding of structural
changes of layered cathode materials during high rate charging/discharging
and the thermal stability during heating are essential in developing new
materials and improving current materials. Since structural changes take
place from the atomic level to the whole electrode level, combination of
characterization techniques covering multilength scales is quite important. In many cases, this means using comprehensive tools
involving diffraction, spectroscopy, and imaging to differentiate the surface from the bulk and to obtain structural/chemical
information with different levels of spatial resolution.
For example, hard X-ray spectroscopy can yield the bulk information and soft X-ray spectroscopy can give the surface
information; X-ray based imaging techniques can obtain spatial resolution of tens of nanometers, and electron-based microcopy
can go to angstroms. In addition to challenges associated with different spatial resolution, the dynamic nature of structural
changes during high rate cycling and heating requires characterization tools to have the capability of collecting high quality data
in a time-resolved fashion. Thanks to the advancement in synchrotron based techniques and high-resolution electron microscopy,
high temporal and spatial resolutions can now be achieved. In this Account, we focus on the recent works studying kinetic and
thermal properties of layer-structured cathode materials, especially the structural changes during high rate cycling and the thermal
stability during heating. Advanced characterization techniques relating to the rate capability and thermal stability will be
introduced. The different structure evolution behavior of cathode materials cycled at high rate will be compared with that cycled
at low rate. Different response of individual transition metals and the inhomogeneity in chemical distribution will be discussed.
For the thermal stability, the relationship between structural changes and oxygen release will be emphatically pointed out. In all
these studies being reviewed, advanced characterization techniques are critically applied to reveal complexities at multiscale in
layer-structured cathode materials.

1. INTRODUCTION life.1−3 However, for large-scale commercialization of EVs, we

Global climate change and environment concerns have
accelerated worldwide research on renewable energy sources
to reduce the use of fossil energy. Rechargeable batteries are the
key technologies for such renewable energies. Among the
efforts for reducing fossil fuel usage in transportation, the
development of various types of electric vehicles (EVs) is a very
important part of it and the rechargeable battery is the key
technology. The lithium-ion battery (LIB) is the most
promising system with high energy density and long cycling
still face many great challenges for future LIBs:4,5 in order to
have longer driving range, the energy density needs to be
significantly improved; the cycling and calendar life need to be
extended, and the cost needs to be lowered; in order to meet
consumer demands for fast charge, the batteries need to have
high rate capability for charging and discharging, especially for
Received: October 10, 2017
Published: January 19, 2018

© 2018 American Chemical Society 290 DOI: 10.1021/acs.accounts.7b00506

Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
charging. In addition, safety characteristic is a very important
issue, since the size of the battery for EV is several orders of
magnitude larger than those for consumer electronics and could
cause disastrous consequences if it catches fire. Unfortunately,
most of the cathode materials used in LIBs today release highly
reactive oxygen at elevated temperatures, which could react
with the carbon and electrolyte in the battery cells to cause
dangerous thermal runaway.4,6 Therefore, improving the
thermal stability of cathode materials against oxygen release is
critically important for safety. In-depth understanding of the
structural changes and their effects on rate capability during
high rate cycling and on thermal stability during heating will
provide valuable information for new material design, current
material improvement, and preventive procedure development
during operations for LIBs. The structural changes during
heating and high rate cycling are dynamic, not static, and fast
data collection capability for characterization tools is a must
have. In addition, since the structural changes take place in a
wide range of length scale,7,8 the combination of character-
ization techniques covering multilength scales from the
electrode level,9 to the particle level10 and the grain and grain
boundary level,11,12 and all the way to the atomic level is
desirable.13 Advances in all kinds of synchrotron based
techniques, such as time-resolved X-ray diffraction (TR-
XRD), time-resolved X-ray absorption (TR-XAS), and trans-
mission X-ray microscopy (TXM), have provided powerful
tools with fast data collection capability for doing kinetic
studies using both spectroscopy and imaging at various levels of
length scales. However, in order to reach atomic level
resolution, transmission electron microscopy (TEM) and
scanning transmission electron microscopy (STEM) techniques
are needed. A schematic illustration is shown in Figure 1,
Figure 1. Schematic illustration of challenges associated with
characterizing cathode materials of Li-ion batteries for vehicle
applications, as well as techniques using (synchrotron based) X-ray
photon and electron beam.
indicating challenges associated with characterizing layered
cathode materials for vehicle application and how synchrotron
X-ray based and electron beam based techniques can help to
address these challenges.
This Account will focus on the review of recent kinetic and
thermal studies on layer-structured transition metal oxide
cathode materials, especially the LiNixMnyCozO2 (x + y + z = 1,
called NMC) materials,14,15 as well as their derivatives, the Li-
rich NMC materials,16,17 since their applications in LIBs have
been expanded rapidly due to their higher energy density than
291
Article
lithium iron phosphate (LFP)18 and lower cost than LiCoO2
materials.19
2. STRUCTURE CHANGES DURING HIGH RATE
CHARGE−DISCHARGE CYCLING
The fundamental understanding of the structure evolution of
electrode materials during the fast charge−discharge process,
especially during fast charge is very important for designing
battery systems for high power applications, such as in hybrid
electric vehicles and pure electric vehicles. However, most of
the research results reported in the literature are focused on
relatively slow processes near the thermodynamic equilibrium
state only, not much on the fast cycling kinetics, mainly due to
the lack of fast data collection techniques with high signal-to-
noise ratio. Thanks to the high flux X-ray beam of the
synchrotron light source, many new time-resolved techniques
have been developed for such dynamic studies. Here are some
examples. Because of the large penetration depth of X-ray
probing beam, X-ray diffraction (XRD) is widely used to obtain
reliable average structural information throughout the whole
sample. Using time-resolved XRD (TR-XRD) technique, Zhou
et al.20 investigated the phase transition behaviors of layered
cathode material LiNi1/3Mn1/3Co1/3O2 (NMC)14 in cells cycled
at various charging rates, especially at high charging rates.
Figure 2 shows the in operando XRD patterns of NMC cathode
collected during the first charge process at current rates of 0.1C,
1C, 10C, 30C, and 60C. During these measurements, data were
collected while the cell was still running. At low current rate of
0.1C, the phase transition process contains two solid solution
reaction regions (hexagonal phase of H1 and H2) at the early
and late stages of charging, and one two-phase coexisting region
(H1 + H2) in the middle stage of charging. Such phase
transition route is different at high current rate as indicated by
in operando XRD with each pattern taking about 4 s to be
collected. When the C rate is increased to 10C, a new broad
peak emerges between H1 and H2 phases and becomes more
and more pronounced at 30C and 60C, indicating the
formation of intermediate phase. In addition, the peak
representing this intermediate phase at 60C charging rate is
much broader than that at low rate, indicating the high
inhomogeneity of Li content in this intermediate phase. At the
end of charge, the intermediate phase is fully converted to the
H2 phase even at the very high rate of 60C. To confirm this
intermediate phase formation observed by TR-XRD, they used
scanning transmission electron microscopy (STEM) to
investigate the structural changes of NMC samples after half
way charging (55 s) at a high rate of 30C, and the results are
shown in Figure 3. STEM has the capability to probe a relevant
location of a sample with very high spatial resolution offering
structural information at atomic scale. A large number of atomic
resolution high-angle annular dark-field (HAADF) and annular-
bright-field (ABF) STEM images were collected. For STEM,
the contrast varies as Z1.7 for the HAADF images, but varies as
Z1/3 for the ABF images where “Z” is the atomic number. Since
HAADF is more sensitive to the heavy atoms, transition metal
(TM) ions are clearly observed from the HAADF image in
bright dots in Figure 3a,b,c, while the lighter atoms of Li and O
are shown more clearly in the ABF image (Figure 3d,e,f). The
uniformly distributed TM layers can be seen from Figure
3a,b,c), indicating that the structure of TM layers is well
preserved after halfway charge at high rate. On the other hand,
the ABF image taken at the same sample position shows two
different types of domains in the green and red dashed squares
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
Figure 2. In operando XRD of NMC during the first charge. Contour plot of the (003) diffraction peak of Li1−xNi1/3Co1/3Mn1/3O2 with increasing x
between x = 0 and 0.7 during the first charge process at different C rates (0.1C, 1C, 10C, 30C, 60C). Data were collected at X14A at NSLS with a
wavelength of 0.7747 Å. Adapted with permission from ref 20. Copyright 2016 John Wiley and Sons.
Figure 3. Structure of half charged NMC at 30C rate. (a) Typical
HAADF image taken along the [110] zone axis of the NMC electrode
after 55 s charging at the current rate of 30C. (b, c) The zoom-in
image of the areas marked with orange and pink squares, respectively.
(d) Corresponding ABF images of NMC electrode after 55 s charging.
(e, f) The zoom-in image of the areas marked with orange and pink
squares, respectively. The blue and green dots indicate the TM ions
and Li ions, respectively. Adapted with permission from ref 20.
Copyright 2016 John Wiley and Sons.
in Figure 3d and in more details in Figure 3e,f. The location of
the Li columns in the green dashed square region in Figure 3e
is different from that in the red dashed square region in Figure
3f. In Figure 3f, the Li columns are no longer located in the
292
Article
middle position of two TM columns (actually, it should be in
the middle of two oxygen columns, but the TM layers have
much stronger contrast). The middle position corresponds to
the typical octahedral sites as shown in the pristine sample. It
looks like the lithium ions in Figure 3f are located at the
pseudotetrahedral sites, as identified by the deviation away from
the middle point of two adjacent TM layers. During the high-
rate charging, the higher overpotential can provide extra driving
force for the motion of lithium ions, causing a largely
inhomogeneous lithium ion distribution and structural defects
within the sample particles. It is assumed that the tetrahedral
site occupation might become energetically favorable for Li in
some Li-poor regions and the intermediate phases are formed
with such occupation. The occurrence of regions with different
contents and occupations of Li distributions might be the origin
of the intermediate phase observed with XRD (as explained in
the previous section). It is likely that intermediate phase formed
during high rate charging serves as a buffer to reduce the local
stress and strain produced by inhomogeneity during charging.
This investigation of intermediate phase formation during high
rate charging for NMC samples was inspired by the previous
reported studies in the literature. Similar intermediate phase
formation during high rate charging was reported in olivine
structured LiFePO4 (LFP) systems.21−23 For example, Liu et
al.22 reported a metastable solid solution phase rather than a
commonly two-phase separation during high-rate cycling by
HR-XRD. Orikasa et al.23 observed a metastable crystal phase
of LixFePO4 during high current charging/discharging, which
contributes to the high-rate performance of LFP system.
Furthermore, Gu et al.24 observed a new intermediate “Li-
staging” phase of Li0.5FePO4 using STEM technique with
atomic spatial resolution. It is worth mentioning that while in
operando XRD and ex situ STEM can sometimes provide
complementary information about the same intermediate
phase, as in layered NMC, it may not be always true in other
situations. For example, for LiFePO4, the intermediate phase
can only last for around 30 min, which makes it difficult to
study using ex situ characterization afterward. Therefore, it is
necessary to first identify the period in which the intermediate
phase can be preserved (if possible, by in situ XRD) before
carrying out ex situ measurements.
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
While XRD provides information on structural change during
electrochemical cycling, X-ray spectroscopy technique, espe-
cially X-ray absorption near edge spectroscopy (XANES) can
identify the contribution of individual transition metals in an
elementally selective way and therefore has been widely used to
study battery materials, which are typically multitransition
metal systems. Despite such wide application, a limited number
of studies on the kinetic characteristics of these transition metal
elements have been reported in the literature. It is very
important to understand the unique contribution of each
transition metal to the material’s kinetic properties for
designing cathode materials with better rate capability. In
operando time-resolved X-ray absorption spectroscopy (TR-
XAS) has the capability to monitor structural changes around
certain transition metal ions in the electrode material during
high rate charge−discharge cycling. Yu et al.25 reported their
studies on the contribution from different transition metals to
the bulk kinetic properties of Li1.2Ni0.15Co0.1Mn0.55O2. This
material belongs to the family of Li-rich Mn-rich layer-
structured materials.16 Many researchers believe that this
material is an intergrowth composite of LiMO2 (M represents
transition metal ions) and Li2MnO3 components.17,26,27 In
order to start the redox reactions of all three transition metals
simultaneously during charge, a 5 V constant-voltage charging
condition, rather than the conventional constant current
charging was applied during the collection of TR-XAS data.
This 5 V charging voltage is higher than all of the Ni2+/Ni3+,
Ni3+/Ni4+, and Co3+/Co4+ redox potentials and the activation
voltage of Li2MnO3. The data collection continued for 900 s
while the 5 V voltage was continuously applied. The data
collection time for each spectrum was optimized to be 15 s to
balance the signal-to-noise ratio and the number of spectra
collected. The extended X-ray absorption fine structure
(EXAFS) part of collected spectra was then Fourier trans-
formed to yield data in R space. The results are shown in a 2D
view in Figure 4.
Figure 4. Five volt constant voltage charging was applied on the
Li1.2Ni0.15Co0.1Mn0.55O2 electrode. Ni, Co, and Mn reacted simulta-
neously, which was recorded using a time-resolved XAS technique.
Projection view of the corresponding Ni−O, Co−O, and Mn−O peak
magnitudes of the Fourier transformed K-edge spectra as functions of
charging time. Adapted with permission from ref 25. Copyright 2014
John Wiley and Sons.
The peaks on the Fourier transformed EXAFS plot
correspond to characteristic bonds within the nearest shells.
The peak position in the plot is usually 0.3−0.4 Å shorter than
the actual bond distance because of phase shift. In the case of
layered material, the first peak corresponds to transition metal−
oxygen bond, which lies within the first shell around transition
metal. Figure 4 shows that the first coordination peak of Ni
293
Article
(Ni−O) experiences dramatic changes in both position and
intensity within the first 100 s, indicating that most of the
charge compensation occurred at the Ni sites. The peak
intensities decreased first (from 0 s to ∼60 s) due to the Jahn−
Teller distortion caused by the oxidation of Ni2+ to Ni3+, then
turned around to increase with further oxidation of Ni3+ to
Ni4+. The EXAFS features remain unchanged after 160 s,
indicating that the oxidation of Ni2+ to Ni4+ was almost
completed within the first 3 min. Compared to Ni, the
evolution of the local structure around Co and Mn sites caused
by Li+ extraction stretched over a longer period of time as
displayed in Figure 4: the first coordination shell peak (Co−O
and Mn−O) intensities show a continuous decrease.
Specifically, Co−O peak experiences most of the intensity
change within 200 s. After that, no significant Co−O peak
intensity changes can be observed. The evolution of the Mn−O
peak intensity generally follows a similar trend to that of the
Co−O peak, but the changes are continuous over the whole
observation time scale (900 s), indicating much slower
delithiation kinetics around Mn sites.
Recently, a new synchrotron based technique, transmission
X-ray microscopy (TXM), has been developed and applied to
battery material studies, especially for in situ studies. Compared
with XAS, which provides average information over a relatively
larger volume of the sample (illuminated by the X-ray beam),
TXM can focus on a nanometer area with higher spatial
resolution. Compared with TEM, which has atomic resolution,
TXM is able to do in situ experiments with spatial resolution
between XAS and TEM and provide a powerful tool to do in
situ studies at the particle level. In addition, TXM is able to do
chemical and oxidation state distribution mapping over an area
of interest, giving us opportunity to visualize the morphological,
chemical, and oxidation state changes of individual particles.
Using TXM, Xu et al.28 carried out an in situ investigation on a
single LiCoO2 particle in a cell during cycling at different rates.
In these experiments, the cell was stopped after cycling to a
certain state to collect the images. As shown in Figure 5, the cell
was sequentially charged and discharged at different rates
(0.2C, 1C, and 10C) between 3 and 4.6 V. The chemical
mapping based on the oxidation state of Co for a selected
Figure 5. In situ monitoring of the chemical inhomogeneity in a single
particle of LiCoO2 up to 20 cycles. Panel a is the chemical map of the
particle at its pristine state. Panel b is at the charged state (at 4.6 V).
Panels c−e are the chemical maps at discharged state (at 3 V) after the
particle went through cycles of different rates at 1C, 10C, and 0.2C,
respectively. Panel f is the map at the discharged state after 20 cycles at
0.2C. All chemical maps are color coded to the corresponding pie
charts. The red area represents the domains at charged state, and the
green area represents the domains at discharged state. Reproduced
with permission from ref 28. Copyright 2017 American Chemical
Society.
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
particle at the end of each discharge process and the end of 1C
charge process is plotted in Figure 5. Using the XANES spectra
of LiCoO2 powder at fully charged and discharged state as
references, they rated the state of charge (SOC) of each spot in
the particle. It can be seen from these mappings, the
distribution of SOC in this particle is inhomogeneous,
especially when high current rate was used. In an ideal
equilibrium situation, the LiCoO2 particle is supposed to return
to the fully discharged state (marked by green) when cell is
discharged to 3 V. However, there were still some parts of the
particle staying at charged state (marked by red) for all cycling
rates. The portion of particle recovered back to discharged state
gets smaller when the rate gets higher. The recovery percentage
of the particle bounced back to a higher level in following low
rate cycling. The inhomogeneity at particle level is induced by
high rate cycling. It is worth noting that the electrochemical
behavior of this particular particle may not be able to represent
the whole cell, especially in capacity retention. However, the
inhomogeneity at particle level is representative. Furthermore,
the 3D chemical map of a LiCoO2 particle recovered from
cycling (Figure 6) shows a crack in particle associated with the
Figure 6. Ex situ 3D chemical map of a selected particle recovered
from a cell that was cycled at 0.2C and disassembled at 3 V. Panels a, b
and c are the same particle in different viewing angles indicated by the
arrows on the bottom. The chemical components are color coded in
the 3D rendering with the discharged stat shown in green and the
charged state shown in red (the portion of the particle that failed to
return to the discharged state at 3 V). Reproduced with permission
from ref 28. Copyright 2017 American Chemical Society.
inactive domain. The appearance of the crack suggests a close
relationship between morphological defects and chemical
inhomogeneity. Since the large chemical inhomogeneity is
induced by high rate cycling as shown in Figure 6,
morphological degradation of the electrode particle could be
a consequence of high rate charge−discharge cycling. Similar
chemical inhomogeneity and cracks in LiCoO2 particles during
high rate cycling was also observed by other researchers using
different techniques.29 A spatial correlation between Li content
distribution and fracture direction was found. Furthermore,
atomic scale observation of structural evolution in LiCoO2
particle during ultrafast charging was also reported by Gong et
al.30 using TEM.
3. CORRELATIONS BETWEEN STRUCTURAL CHANGES
AND OXYGEN RELEASE DURING HEATING
As mentioned before, studying the structural changes of layered
oxide cathode materials during heating is very important in
providing guidance for the improvement of thermal stability,
which is critically related to the safety characteristics of these
cathode materials. Cathode heating, caused by electric short
circuit, mechanical impact, or other accidental events could
trigger dangerous thermal runaway. During such processes, the
fast temperature increase induces structural changes of cathode
294
Article
material, which is usually accompanied by oxygen release. The
newly released oxygen can react with the carbon and binder in
the cathode, and even with the flammable organic solvent in the
electrolyte, leading to the thermal runaway.31 Therefore, it is
very important to study the correlation between structural
changes and oxygen release of cathode materials during heating
by recording the information about both of them simulta-
neously. Nam et al. reported their results of combined mass
spectroscopy (MS) and X-ray diffraction (XRD) during heating
using a specially designed heating cell at synchrotron facilities.32
The strong X-ray source at synchrotron beamlines and the fast
2D area detector provided high quality time-resolved XRD
data, and the MS provided gas evolution correlating to the
structural changes during heating. Taking advantage of this
technique, studies were carried out on Li0.33Ni0.8Co0.15Al0.05O2
(NCA). It was reported that during heating, two distinct
oxygen-release peaks with their centers around 230 and 300 °C
appeared in the MS results (Figure 7). These peaks are clearly
Figure 7. (a) Time-resolved (TR) XRD patterns and (b) results from
simultaneously measured mass spectroscopy (MS) that trace oxygen
gas released from the overcharged Li0.33Ni0.8Co0.15Al0.05O2 during
heating up to 500 °C. The cathode sample in a glass tube was heated
from 25 to 500 °C in 4 h during the TR-XRD and MS measurements
(heating rate 2.0 °C min−1). The left panel shows the ideal crystal-
structure models for rhombohedral, spinel, and rock-salt structures.
Adapted with permission from ref 32. Copyright 2013 John Wiley and
Sons.
correlated with the two structural phase-transitions shown in
the time-resolved XRD (TR-XRD) patterns in the left panel of
Figure 7. The beginning of first noticeable oxygen release
observed at about 180 °C is close to the starting temperature of
the structural transition from the initial layered structure to the
spinel structure, as shown in the TR-XRD results. After the
intensity of the first peak reached the maximum at about 230
°C and started decreasing, a second oxygen release peak arose
around 250 °C and continued to 500 °C, with its center at
about 300 °C. This second peak relates to the phase transition
from the disordered spinel structure to the NiO-like rock-salt
structure between 250 and 500 °C as shown in the TR-XRD.
More O2 gas was released during the second phase transition
than during the first one.
Direct correlation between structural changes and oxygen
release during the thermal decomposition of
LixNi0.8Co0.15Al0.05O2 cathode materials overcharged to x =
0.5, 0.33, and 0.1 levels was reported by Bak et al.33 This study
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research Article

showed that the state of charge affects both the structural stability. For the unheated overcharged NCA, TEM imaging
changes that occur in these materials and the evolution of O2 (shown in Figure 9) revealed three structures, the layered
and CO2 gases during thermal decomposition. The evolution of
both O2 and CO2 gases are well-correlated with the phase
transitions that occur during thermal decomposition. The
higher the level of overcharging (i.e., overdelithiation) leads to
the poorer the thermal stability and the larger amounts of
oxygen release at lower temperatures. In particular, highly
overcharged cathode samples (i.e., Li0.1Ni0.8Co0.15Al0.05O2)
show a severe oxygen release at a temperature as low as 175
°C, accompanied by a sudden structural change from the
layered to the disordered spinel phase.
Bak et al. also reported their systematic studies on the
correlations between the structural changes and oxygen release
for LiNixMnyCozO2 (NMC, x + y + z = 1) materials with
different Co/Ni ratios (NMC433, NMC532, NMC622, and Figure 9. HRTEM images viewed along (a) [1̅2̅1]R and (b) [001]R
NMC811).34 As shown in Figure 8, the effects of the Co/Ni directions for an overcharged Li0.15Ni0.8Co0.15Al0.05O2 particle. The
insets in panel a show the magnified images from the small rectangle
areas, indicating the rock-salt (RS), spinel (S), and rhombohedral (R)
structures. In panel b, although both the spinel and rock-salt structures
have hexagonal pattern, the periodicity of the former is twice as that of
the latter. Reproduced with permission from ref 32. Copyright 2013
John Wiley and Sons.

(located in the core, far from the surface), the spinel (the shell
near the surface), and the rock-salt (at the surface layer or edge
of the particle). Figure 10 shows the in situ TEM results of
overcharged NCA, indicating that upon heating, spinel phase
(shell) propagates toward the core of the particle and the rock-
salt phase (surface) grows toward the spinel phase region
(shell). For NMC333, the overcharged particles also have
layered phase in the core, the spinel phase in the shell.
However, the great difference is, rather than the small amount

Figure 8. Mass spectroscopy profiles for the oxygen (O2, m/z = 32)
collected simultaneously during measurement of TR-XRD and the
corresponding temperature region of the phase transitions for NMC
samples (lower panel). Reproduced with permission from ref 34.
Copyright 2014 American Chemical Society.

ratio on the thermal structural stability can be clearly seen. For

NMC422 and NMC532 samples with higher Co/Ni ratios, the
phase transition to both LiMn2O4 type and Co3O4 type spinels
occurred at higher temperatures and the Co3O4 type spinel
remained as the dominating phase all the way up to 600 °C. In
contrast, for the NMC622 and NMC811 samples with lower
Co/Ni ratios, the phase transitions to both LiMn2O4 type and
Co3O4 type spinels started and finished at much lower
temperatures, and the phase transition to rock-salt phase was
completed at 550 and 365 °C, respectively.
The structural changes and oxygen release differences
between NCA and NMC agreed very well with an earlier
study about the phase distribution of them at the particle level
using in situ transmission electron microscopy (TEM) during
Figure 10. HRTEM images taken from an overcharged
heating. Wu et al. used high resolution TEM (HRTEM) to Li0.15Ni0.8Co0.15Al0.05O2 particle (a) before heating, and after heating
study highly delithiated NCA and NMC333 and found that at (b)100 °C, (c) 200 °C, and (d) 300 °C. The insets are the
their particles both feature the core−shell−surface structure corresponding SAEDPs taken from a large area (approximately 240
after overcharging (over delithiation).35 However, the detailed nm in diameter) of the same particle. RS, rock-salt; S, spinel; R,
phase distributions differ significantly between NCA and NMC, rhombohedral. Reproduced with permission from ref 32. Copyright
which has direct consequence on their respective thermal 2013 John Wiley and Sons.

295 DOI: 10.1021/acs.accounts.7b00506

Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
of rock-salt phase at the surface in NCA, small amount of O1
phase (CdI2 type structure featuring hexagonal close-packing of
oxygen, rather than cubic close-packing of oxygen as in layered,
spinel, and rock salt) was observed on the surface. Upon
heating, the surface O1 phase transforms into a Co3O4 type
spinel and the shell spinel phase propagates toward the core.
This O1 to Co3O4 type spinel phase transition pushed the
spinel to rock salt transition to a very high temperature,
contributing to the good thermal stability of NMC333. It is
interesting to note that O1 phase is usually formed when
layered material is highly delithiated (e.g., with 0.9Li
extracted).36,37 The fact that layered phase being the core,
spinel phase being the shell and O1 phase being the surface
suggests the considerable structural inhomogeneity within the
particle. As a result, such inhomogeneity induced during the
electrochemical process can greatly influence the thermal
stability of the material.
Aside from reasons mentioned above, different kinetic and
thermodynamic properties of transition metal elements also
contribute to the difference on thermal stability between
NMC333 and NCA. For the first step of phase transition (from
layered to LiMn2O4-type spinel), theoretical calculation shows
that while the layered to spinel phase transition is always
thermodynamically driven (for delithiated cathode), at what
specific temperature such transition takes place is really a
kinetic issue.38 This is because it requires the migration of
transition metal cations from the original octahedral sites in the
transition metal layer to the octahedral sites in the adjacent
lithium layer. It turned out such migration is in fact via the
intermediate tetrahedral site as this path is the most
energetically favorable one. Based on this consideration,
octahedral site stabilization energy (OSSE), which calculates
the energy difference between octahedral and tetrahedral
occupation is a good indicator of the mobility of transition
metal cations.39 Figure 11 shows detailed OSSE of various
Figure 11. Tendency of migration (based on octahedral site
stabilization energy) as a function of the electronic structure of 3d
transition metal cations. Exchange energy is not considered in
calculating the OSSE. Reproduced with permission from ref 39.
Copyright 2016 John Wiley and Sons.
cations. It indicates that OSSE is highly dependent on specific
electronic structure, which is determined by the oxidization
state and spin state. From this aspect, the better thermal
stability of NMC in comparison to NCA can be attributed to
the different transition metal composition. For delithiated
NMC333, there are large amounts of Co3+ (low spin), Co4+
(low spin), and Mn4+, all of which are highly stable in
octahedral sites (indicated by the fairly negative OSSE).
Therefore, the energy barrier of migrating to the intermediate
tetrahedral site is high and the layered to spinel transition is
296
Article
kinetically sluggish. This not only pushes the phase transition to
occur at a higher temperature but also kinetically stabilizes the
LiMn2O4 type phase over a large temperature range. The role
of Co and Mn are further confirmed by thermal stability study
of a series of NMC materials with varying element ratios. It
shows that the thermal stability of NMC materials improves as
the concentration of Co and Mn increase. The benefit of having
Co and Mn (especially Co) in the material also extends to
higher temperature when the second step reaction is initiated.
At high temperature, the cations may become mobile enough
that kinetics is no longer an issue. The reaction route is mainly
determined by thermodynamics or by the phase diagram. For
the Li−Ni−O2 phase diagram, NiO rock salt is a very stable
phase, and for the Li−Co−O2 phase diagram, Co3O4 spinel is
the stable one.40 Consequently, nickel-rich materials tend to
form rock-salt phase at high temperature and cobalt-rich
materials tend to form Co3O4 type spinel. Because Co3O4
formation leads to less oxygen release than rock-salt formation,
adding Co to layered material helps to improve the thermal
stability.
4. CONCLUDING REMARKS
The high rate charge/discharge capability and safety character-
istics are two of the main challenges to be addressed for the
large-scale LIB application for electric vehicles. In this Account,
we have reviewed major characterization techniques in
multilength scales for monitoring the structural evolution and
kinetic characteristics of various transition metal elements for
layer-structured transition metal oxides such as LiNixMnyCozO2
(x + y + z = 1, called NMC), Li-rich materials, and LiCoO2
during the fast charge/discharge processes. Formation of the
intermediate phase during high rate cycling was detected by
combination of TR-XRD and STEM in LiNi1/3Mn1/3Co1/3O2,
which can serve as a buffer to reduce the local stress and strain
produced by inhomogeneity during charging. In Li-rich
material, much slower kinetics of Mn than Co and Ni during
quick charge/discharge process was monitored by TR-XAS and
TEM. In addition, the chemical inhomogeneity and cracks on
the particle level of LiCoO2 were observed by in situ TXM. We
have also summarized the advanced characterization methods
for studying the correlation between structural changes and
oxygen release of layered cathode materials during heating. The
crucial insights obtained from TR-XRD and in situ TEM show
that the thermal stability of the cathode materials depends on
the transition metal chemistry and phase inhomogeneity at the
particle level. The advanced characterization techniques
covering multilength scales play important roles in guiding
the optimization of current layered electrode materials, as well
as in designing novel cathode materials with high power
capability and good thermal stability.
■ AUTHOR INFORMATION
Corresponding Authors
*X.Y. E-mail: xyu@iphy.ac.cn.
*X.-Q.Y. E-mail: xyang@bnl.gov.
ORCID
Xiqian Yu: 0000-0001-8513-518X
Xiao-Qing Yang: 0000-0002-3625-3478
Author Contributions
§
E. Hu and X. Wang contributed equally.
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research
Notes
The authors declare no competing financial interest.
Biographies
Enyuan Hu is a postdoctoral researcher at the Chemistry Division of
Brookhaven National Laboratory. His research focuses on lithium-ion
batteries.
Xuelong Wang is a Ph.D. candidate at Institute of Physics, Chinese
Academy of Sciences. His research focuses on first-principle calculation
and simulation of lithium-ion battery materials.
Xiqian Yu is an associate professor at Institute of Physics, Chinese
Academy of Sciences. His research interest is focused on the
characterization of the electrode materials for rechargeable batteries.
Xiao-Qing Yang is the group leader of Electrochemical Energy
Storage research group at Chemistry Division of Brookhaven National
Laboratory. His research group focuses on material research of
advanced batteries, especially on batteries for vehicle technology.
■ ACKNOWLEDGMENTS
The work done at Brookhaven National Laboratory was
supported by the Assistant Secretary for Energy Efficiency and
Renewable Energy, Office of Vehicle Technologies of the U.S.
Department of Energy through the Advanced Battery Materials
Research (BMR) Program, including Battery500 Consortium
under Contract DE-SC0012704. The work at IOP was
supported by funding from “One Hundred Talent Project” of
the Chinese Academy of Sciences and Ministry of Science and
Technology of China (Grants 2016YFB0100300).
■
REFERENCES
(1) Tarascon, J. M.; Armand, M. Issues and challenges facing
rechargeable lithium batteries. Nature 2001, 414, 359−367.
(2) Goodenough, J. B.; Park, K.-S. The Li-ion rechargeable battery: a
perspective. J. Am. Chem. Soc. 2013, 135, 1167−1176.
(3) Whittingham, M. S. Lithium batteries and cathode materials.
Chem. Rev. 2004, 104, 4271−4301.
(4) Lu, L.; Han, X.; Li, J.; Hua, J.; Ouyang, M. A review on the key
issues for lithium-ion battery management in electric vehicles. J. Power
Sources 2013, 226, 272−288.
(5) Smith, K.; Wang, C.-Y. Power and thermal characterization of a
lithium-ion battery pack for hybrid-electric vehicles. J. Power Sources
2006, 160, 662−673.
(6) Balakrishnan, P.; Ramesh, R.; Kumar, T. P. Safety mechanisms in
lithium-ion batteries. J. Power Sources 2006, 155, 401−414.
(7) Abraham, A.; Housel, L. M.; Lininger, C. N.; Bock, D. C.; Jou, J.;
Wang, F.; West, A. C.; Marschilok, A. C.; Takeuchi, K. J.; Takeuchi, E.
S. Investigating the complex chemistry of functional energy storage
systems: the need for an integrative, multiscale (molecular to
mesoscale) perspective. ACS Cent. Sci. 2016, 2, 380−387.
(8) Lin, F.; Markus, I. M.; Nordlund, D.; Weng, T.-C.; Asta, M. D.;
Xin, H. L.; Doeff, M. M. Surface reconstruction and chemical
evolution of stoichiometric layered cathode materials for lithium-ion
batteries. Nat. Commun. 2014, 5, 3529.
(9) Shearing, P.; Howard, L.; Jørgensen, P. S.; Brandon, N.; Harris, S.
Characterization of the 3-dimensional microstructure of a graphite
negative electrode from a Li-ion battery. Electrochem. Commun. 2010,
12, 374−377.
(10) Lin, F.; Nordlund, D.; Li, Y.; Quan, M. K.; Cheng, L.; Weng, T.-
C.; Liu, Y.; Xin, H. L.; Doeff, M. M. Metal segregation in hierarchically
structured cathode materials for high-energy lithium batteries. Nat.
Energy 2016, 1, 15004.
(11) Hu, E. Y.; Lyu, Y. C.; Xin, H. L. L.; Liu, J.; Han, L. L.; Bak, S.
M.; Bai, J. M.; Yu, X. Q.; Li, H.; Yang, X. Q. Explore the Effects of
297
Article
Microstructural Defects on Voltage Fade of Li- and Mn-Rich
Cathodes. Nano Lett. 2016, 16, 5999−6007.
(12) Yan, P.; Zheng, J.; Gu, M.; Xiao, J.; Zhang, J.-G.; Wang, C.-M.
Intragranular cracking as a critical barrier for high-voltage usage of
layer-structured cathode for lithium-ion batteries. Nat. Commun. 2017,
8, 14101.
(13) Xu, B.; Fell, C. R.; Chi, M.; Meng, Y. S. Identifying surface
structural changes in layered Li-excess nickel manganese oxides in high
voltage lithium ion batteries: A joint experimental and theoretical
study. Energy Environ. Sci. 2011, 4, 2223−2233.
(14) Yabuuchi, N.; Ohzuku, T. Novel lithium insertion material of
LiCo1/3Ni1/3Mn1/3O2 for advanced lithium-ion batteries. J. Power
Sources 2003, 119, 171−174.
(15) Delmas, C.; Menetrier, M.; Croguennec, L.; Saadoune, I.;
Rougier, A.; Pouillerie, C.; Prado, G.; Grüne, M.; Fournes, L. An
overview of the Li(Ni, M)O2 systems: syntheses, structures and
properties. Electrochim. Acta 1999, 45, 243−253.
(16) Lu, Z.; MacNeil, D.; Dahn, J. Layered cathode materials
Li[NixLi(1/3−2x/3)Mn(2/3−x/3)]O2 for lithium-ion batteries. Electrochem.
Solid-State Lett. 2001, 4, A191−A194.
(17) Thackeray, M. M.; Johnson, C. S.; Vaughey, J. T.; Li, N.;
Hackney, S. A. Advances in manganese-oxide ’composite’ electrodes
for lithium-ion batteries. J. Mater. Chem. 2005, 15, 2257−2267.
(18) Padhi, A. K.; Nanjundaswamy, K.; Goodenough, J. B. Phospho-
olivines as positive-electrode materials for rechargeable lithium
batteries. J. Electrochem. Soc. 1997, 144, 1188−1194.
(19) Mizushima, K.; Jones, P.; Wiseman, P.; Goodenough, J. B.
LixCoO2 (0<x≤1): A new cathode material for batteries of high energy
density. Mater. Res. Bull. 1980, 15, 783−789.
(20) Zhou, Y.-N.; Yue, J.-L.; Hu, E.; Li, H.; Gu, L.; Nam, K.-W.; Bak,
S.-M.; Yu, X.; Liu, J.; Bai, J.; Dooryhee, E.; Fu, Z.-W.; Yang, X.-Q.
High-Rate Charging Induced Intermediate Phases and Structural
Changes of Layer-Structured Cathode for Lithium-Ion Batteries. Adv.
Energy Mater. 2016, 6, 1600597.
(21) Lim, J.; Li, Y. Y.; Alsem, D. H.; So, H.; Lee, S. C.; Bai, P.;
Cogswell, D. A.; Liu, X. Z.; Jin, N.; Yu, Y. S.; Salmon, N. J.; Shapiro, D.
A.; Bazant, M. Z.; Tyliszczak, T.; Chueh, W. C. Origin and hysteresis
of lithium compositional spatiodynamics within battery primary
particles. Science 2016, 353, 566−571.
(22) Liu, H.; Strobridge, F. C.; Borkiewicz, O. J.; Wiaderek, K. M.;
Chapman, K. W.; Chupas, P. J.; Grey, C. P. Capturing metastable
structures during high-rate cycling of LiFePO4 nanoparticle electrodes.
Science 2014, 344, 1252817.
(23) Orikasa, Y.; Maeda, T.; Koyama, Y.; Murayama, H.; Fukuda, K.;
Tanida, H.; Arai, H.; Matsubara, E.; Uchimoto, Y.; Ogumi, Z. Direct
Observation of a Metastable Crystal Phase of LixFePO4 under
Electrochemical Phase Transition. J. Am. Chem. Soc. 2013, 135,
5497−5500.
(24) Gu, L.; Zhu, C. B.; Li, H.; Yu, Y.; Li, C. L.; Tsukimoto, S.; Maier,
J.; Ikuhara, Y. Direct Observation of Lithium Staging in Partially
Delithiated LiFePO4 at Atomic Resolution. J. Am. Chem. Soc. 2011,
133, 4661−4663.
(25) Yu, X. Q.; Lyu, Y. C.; Gu, L.; Wu, H. M.; Bak, S. M.; Zhou, Y.
N.; Amine, K.; Ehrlich, S. N.; Li, H.; Nam, K. W.; Yang, X. Q.
Understanding the Rate Capability of High-Energy-Density Li-Rich
Layered Li1.2Ni0.15Co0.1Mn0.55O2 Cathode Materials. Adv. Energy
Mater. 2014, 4, 1300950.
(26) Johnson, C. S.; Kim, J. S.; Lefief, C.; Li, N.; Vaughey, J. T.;
Thackeray, M. M. The significance of the Li2MnO3 component in
’composite’ xLi2MnO3·(1-x)LiMn0.5Ni0.5O2 electrodes. Electrochem.
Commun. 2004, 6, 1085−1091.
(27) Long, B. R.; Croy, J. R.; Dogan, F.; Suchomel, M. R.; Key, B.;
Wen, J. G.; Miller, D. J.; Thackeray, M. M.; Balasubramanian, M. Effect
of Cooling Rates on Phase Separation in 0.5Li2MnO3·0.5LiCoO2
Electrode Materials for Li-Ion Batteries. Chem. Mater. 2014, 26,
3565−3572.
(28) Xu, Y. H.; Hu, E. Y.; Zhang, K.; Wang, X. L.; Borzenets, V.; Sun,
Z. H.; Pianetta, P.; Yu, X. Q.; Liu, Y. J.; Yang, X. Q.; Li, H. In situ
Visualization of State-of-Charge Heterogeneity within a LiCoO2
DOI: 10.1021/acs.accounts.7b00506
Acc. Chem. Res. 2018, 51, 290−298
Accounts of Chemical Research Article

Particle that Evolves upon Cycling at Different Rates. ACS Energy Lett.
2017, 2, 1240−1245.
(29) Diercks, D. R.; Musselman, M.; Morgenstern, A.; Wilson, T.;
Kumar, M.; Smith, K.; Kawase, M.; Gorman, B. P.; Eberhart, M.;
Packard, C. E. Evidence for Anisotropic Mechanical Behavior and
Nanoscale Chemical Heterogeneity in Cycled LiCoO2. J. Electrochem.
Soc. 2014, 161, F3039−F3045.
(30) Gong, Y.; Zhang, J. N.; Jiang, L. W.; Shi, J. A.; Zhang, Q. H.;
Yang, Z. Z.; Zou, D. L.; Wang, J. Y.; Yu, X. Q.; Xiao, R. J.; Hu, Y. S.;
Gu, L.; Li, H.; Chen, L. Q. In Situ Atomic-Scale Observation of
Electrochemical Delithiation Induced Structure Evolution of LiCoO2
Cathode in a Working All-Solid-State Battery. J. Am. Chem. Soc. 2017,
139, 4274−4277.
(31) MacNeil, D. D.; Lu, Z. H.; Chen, Z. H.; Dahn, J. R. A
comparison of the electrode/electrolyte reaction at elevated temper-
atures for various Li-ion battery cathodes. J. Power Sources 2002, 108,
8−14.
(32) Nam, K. W.; Bak, S. M.; Hu, E. Y.; Yu, X. Q.; Zhou, Y. N.;
Wang, X. J.; Wu, L. J.; Zhu, Y. M.; Chung, K. Y.; Yang, X. Q.
Combining In Situ Synchrotron X-Ray Diffraction and Absorption
Techniques with Transmission Electron Microscopy to Study the
Origin of Thermal Instability in Overcharged Cathode Materials for
Lithium-Ion Batteries. Adv. Funct. Mater. 2013, 23, 1047−1063.
(33) Bak, S. M.; Nam, K. W.; Chang, W.; Yu, X. Q.; Hu, E. Y.;
Hwang, S.; Stach, E. A.; Kim, K. B.; Chung, K. Y.; Yang, X. Q.
Correlating Structural Changes and Gas Evolution during the Thermal
Decomposition of Charged LixNi0.8Co0.15Al0.05O2 Cathode Materials.
Chem. Mater. 2013, 25, 337−351.
(34) Bak, S. M.; Hu, E. Y.; Zhou, Y. N.; Yu, X. Q.; Senanayake, S. D.;
Cho, S. J.; Kim, K. B.; Chung, K. Y.; Yang, X. Q.; Nam, K. W.
Structural Changes and Thermal Stability of Charged LiNixMnyCozO2
Cathode Materials Studied by Combined In Situ Time-Resolved XRD
and Mass Spectroscopy. ACS Appl. Mater. Interfaces 2014, 6, 22594−
22601.
(35) Wu, L. J.; Nam, K. W.; Wang, X. J.; Zhou, Y. N.; Zheng, J. C.;
Yang, X. Q.; Zhu, Y. M. Structural Origin of Overcharge-Induced
Thermal Instability of Ni-Containing Layered-Cathodes for High-
Energy-Density Lithium Batteries. Chem. Mater. 2011, 23, 3953−3960.
(36) Amatucci, G. G.; Tarascon, J. M.; Klein, L. C. CoO2, the end
member of the LixCoO2 solid solution. J. Electrochem. Soc. 1996, 143,
1114−1123.
(37) Das, H.; Urban, A.; Huang, W.; Ceder, G. First-principles
simulation of the (Li-Ni-Vacancy) O phase diagram and its relevance
for the surface phases in Ni-rich Li-ion cathode materials. Chem. Mater.
2017, 29, 7840−7851.
(38) Reed, J.; Ceder, G. Role of electronic structure in the
susceptibility of metastable transition-metal oxide structures to
transformation. Chem. Rev. 2004, 104, 4513−4533.
(39) Hu, E. Y.; Bak, S. M.; Liu, Y. J.; Liu, J.; Yu, X. Q.; Zhou, Y. N.;
Zhou, J. G.; Khalifah, P.; Ariyoshi, K.; Nam, K. W.; Yang, X. Q.
Utilizing Environmental Friendly Iron as a Substitution Element in
Spinel Structured Cathode Materials for Safer High Energy Lithium-
Ion Batteries. Adv. Energy Mater. 2016, 6, 1501662.
(40) Wang, L.; Maxisch, T.; Ceder, G. A first-principles approach to
studying the thermal stability of oxide cathode materials. Chem. Mater.
2007, 19, 543−552.

298 DOI: 10.1021/acs.accounts.7b00506

Acc. Chem. Res. 2018, 51, 290−298