Thermally driven interfacial degradation between Li7La3Zr2O12 electrolyte and LiNi0.6Mn0.2Co0.2O2 cathode

Article abstract of DOI:10.1021/acs.chemmater.0c02261

Solid-state batteries offer higher energy density and enhanced safety compared to the present lithium-ion batteries using liquid electrolytes. A challenge to implement them is the high resistances, especially at the solid electrolyte interface with the cathode. Sintering at elevated temperature is needed in order to get good contact between the ceramic solid electrolyte and oxide cathodes and thus to reduce contact resistances. Many solid electrolyte and cathode materials react to form secondary phases. It is necessary to find out which phases arise as a result of interface sintering and evaluate their effect on electrochemical properties. In this work, we assessed the interfacial reactions between LiNi_0.6Mn_0.2Co_0.2O₂ (NMC622) and Li₇La₃Zr₂O₁₂ (LLZO) as a function of temperature in air. We prepared model systems by depositing thin-film NMC622 cathode layers on LLZO pellets. The thin-film cathode approach enabled us to use interface-sensitive techniques such as X-ray absorption spectroscopy in the near-edge as well as the extended regimes and identify the onset of detrimental reactions. We found that the Ni and Co chemical environments change already at moderate temperatures, on-setting from 500 °C and becoming especially prominent at 700 °C. By analyzing spectroscopy results along with X-ray diffraction, we identified Li₂CO₃, La₂Zr₂O₇, and La(Ni,Co)O₃ as the secondary phases that formed at 700 °C. The interfacial resistance for Li transfer, measured by electrochemical impedance spectroscopy, increases significantly upon the onset and evolution of the detected interface chemistry. Our findings suggest that limiting the bonding temperature and avoiding CO₂ in the sintering environment can help to remedy the interfacial degradation.

Full text of DOI:10.1021/acs.chemmater.0c02261

pubs.acs.org/cm
Article
Thermally Driven Interfacial Degradation between Li7La3Zr2O12
Electrolyte and LiNi0.6Mn0.2Co0.2O2 Cathode
̈
Younggyu Kim, Dongha Kim, Roland Bliem, Gulin Vardar, Iradwikanari Waluyo, Adrian Hunt,
Joshua T. Wright, John P. Katsoudas, and Bilge Yildiz*
Cite This: https://dx.doi.org/10.1021/acs.chemmater.0c02261
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Solid-state batteries offer higher energy density and
enhanced safety compared to the present lithium-ion batteries using
liquid electrolytes. A challenge to implement them is the high
resistances, especially at the solid electrolyte interface with the
cathode. Sintering at elevated temperature is needed in order to get
good contact between the ceramic solid electrolyte and oxide
cathodes and thus to reduce contact resistances. Many solid
electrolyte and cathode materials react to form secondary phases.
It is necessary to find out which phases arise as a result of interface
sintering and evaluate their effect on electrochemical properties. In
this work, we assessed the interfacial reactions between
LiNi0.6Mn0.2Co0.2O2 (NMC622) and Li7La3Zr2O12 (LLZO) as a
function of temperature in air. We prepared model systems by
depositing thin-film NMC622 cathode layers on LLZO pellets. The
thin-film cathode approach enabled us to use interface-sensitive techniques such as X-ray absorption spectroscopy in the near-edge
as well as the extended regimes and identify the onset of detrimental reactions. We found that the Ni and Co chemical environments
change already at moderate temperatures, on-setting from 500 °C and becoming especially prominent at 700 °C. By analyzing
spectroscopy results along with X-ray diffraction, we identified Li2CO3, La2Zr2O7, and La(Ni,Co)O3 as the secondary phases that
formed at 700 °C. The interfacial resistance for Li transfer, measured by electrochemical impedance spectroscopy, increases
significantly upon the onset and evolution of the detected interface chemistry. Our findings suggest that limiting the bonding
temperature and avoiding CO2 in the sintering environment can help to remedy the interfacial degradation.
In contrast, progress at the cathode|LLZO interface remains
relatively slow. Computational studies based on first-principles
and thermodynamic calculations have predicted that the
interface of layered oxide cathodes with the LLZO interface
is more stable compared to other interfaces that comprise
sulfide electrolytes (Li10GeP2S12 and Li3PS4) or other types of
cathodes (LiMn2O4 and LiFePO4).13,14 However, the actual
demonstration of all solid batteries using layered oxide cathode
and LLZO has been dissatisfactory so far. For example, cells
comprising LiCoO2 (LCO) and LLZO sintered at 700 °C
suffered from poor capacity (35 mAh/g).15 The main issue
here arises from the secondary-phase formation at the cathode|
LLZO interface during heat treatment, which leads to high
interfacial resistance. Even though there have been attempts to
1. INTRODUCTION
Current lithium-ion batteries using liquid electrolytes are
inherently unstable because of low flash points of carbonate
species.1 Substituting the liquid electrolyte with a solid
electrolyte may solve this problem and obtain high-capacity,
high-power cells without flammability issues.2,3 Among the
possible choices for the solid electrolyte, Li7La3Zr2O12 (LLZO)
is a promising material system because of its relatively high
conductivity, which reaches up to 10−1 to 1 mS/cm.4 In
addition, its large electrochemical window permits the usage of
both a high-voltage cathode and the Li metal anode.5
Exploiting Li metal anode could lead to the development of
batteries with much higher capacity than the current lithium-
ion batteries using graphite as an anode.6 Even with solid
electrolytes such as LLZO, Li dendrites can grow through grain
boundaries.7,8 However, there are many promising recent
developments to overcome this problem.9−11 In addition, there
has been much progress on the Li wettability on LLZO by
removing the Li2CO3 layer, and this has already lead to
significantly lowered interfacial resistance at the LLZO|Li
interface.12
Received: May 29, 2020
Revised: October 5, 2020
https://dx.doi.org/10.1021/acs.chemmater.0c02261
© XXXX American Chemical Society
Chem. Mater. XXXX, XXX, XXX−XXX
A

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 1. (a) O K-edge X-ray absorption spectra (partial fluorescence yield mode) for 100 nm NMC622|LLZO in the as-deposited state and
annealed at different temperatures in air for 4 h. (b) Region A normalized by highest point in each region. Data shifted by arbitrarily chosen offset
for comparison.
characterize chemical reactions between layered oxide
cathodes and LLZO, a consensus on the nature of the
decomposition products is missing,16−18 and the onset
conditions for the degradation remain unexplored. Further-
more, theoretical studies have predicted the formation of Li-
rich phases such as Li2NiO3, Li5CoO4, and Li2MnO3,13,19 but
they have not predicted phases that could form between the
LLZO and the layered oxide cathode. This prediction does not
match well with experimental findings in the literature15−17,20
as well as in our recent21 and present work. We believe that the
discrepancy comes from the role of different chemical species
and elevated temperatures in the environment, which were not
considered in those calculations, and as discussed later in this
paper.
Probing the cathode−solid electrolyte interface is a major
challenge. X-ray diffraction (XRD) on mixed powders is a
practical approach to identify reaction phases between solid
electrolyte and cathode materials. However, for detecting the
secondary phases with good sensitivity using this bulk
approach, the reaction has to proceed significantly within the
detection limit of powder XRD. Thus, it misses the ability to
detect the onset conditions and identify the effect of very thin
reaction zones on the charge-transfer kinetics. We have
recently addressed this challenge by using thin-film cathode
layers as model systems deposited on solid electrolyte pellets.21
This approach positions the interfacial region within the
detection depth of techniques such as X-ray absorption
spectroscopy (XAS). This enabled us to capture the initiation
of the degradation process at relatively low temperatures and
to precisely characterize the secondary phases and the reacted
zone depth (on the order of nanometer scale) at the LiCoO2|
LLZO interface.21
operating voltage and capacity,24 and this makes the system
more interesting for industrial applications.
Our approach combines synchrotron techniques [X-ray
absorption near-edge spectroscopy (XANES) and extended X-
ray absorption fine structure (EXAFS)] and lab-based
techniques [XRD and electrochemical impedance spectroscopy
(EIS)]. XRD enables preliminary phase analysis, but XRD
alone is not sufficient to resolve the onset of the degradation
where phases have low crystallinity and overlapping peaks. We
used XANES and EXAFS to characterize the local chemical
and atomic coordination environment, even at relatively low
temperatures, where the onset of reactions has taken place but
the products have not formed long-range order yet. By
combining the results obtained from our work and former
literature on the instability of layered oxides, we explain the
degradation behavior at the NMC622|LLZO interface. Results
showed that cation intermixing and oxidation starts at 300 °C,
and the effect becomes more severe as temperature rises in air.
We found formation and crystallization of Li2CO3 starting
from 500 °C. At 700 °C, La(Ni,Co)O3 and La2Zr2O7 were also
detected. We assessed the impact of the chemically degraded
interface on the charge-transfer properties by performing EIS
on symmetric NMC622|LLZO|NMC622 samples. Interfacial
resistance increased with annealing temperature, showing that
chemical degradation at the interface leads to blockage of Li-
ion transfer through the interface.
2. EXPERIMENT DESIGN AND METHODS
In order to understand interfacial degradation between NMC622
cathode and LLZO, it is necessary to keep the interfacial region within
the detection depth of characterization techniques. In our experiment,
XANES experiment using soft X-rays set the upper limit of the
thickness of NMC622 thin films that could be used. The thickness
was chosen considering the mean free path of photons with 1 keV
energy,25 which is 100 nm. Samples were prepared by depositing
NMC622 on top of preprepared LLZO pellets by RF sputtering at
nominally room temperature conditions. After deposition, the samples
were annealed at 300, 500, and 700 °C in air for 4 h. Annealing was
done in air in order to consider the possible effect of CO2 during
chemical reaction. Temperature range was chosen to include the
range for crystallizing layered oxide cathode thin films after
deposition, which spans between 400 and 700 °C.26−29
In this research, we have adopted the same thin-film cathode
approach as in our recent work21 and have focused on the
layered oxide cathode, LiNi0.6Mn0.2Co0.2O2 (NMC622) on the
LLZO electrolyte. By substituting Co with Ni, it is possible to
extract more electrons from the cathode during charging
without triggering oxygen evolution.22 In addition, Mn
incorporation gives additional stability by suppressing Ni4+
formation.23 Using NMC622 could lead to cells with high
B
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 2. (a) La M5, La M4, Ni L3, and Ni L2 X-ray absorption spectra (partial fluorescence yield mode) for 100 nm NMC622|LLZO in the as-
deposited state and annealed at different temperatures in air. Amplitude of peaks normalized by a La M5 peak. (b) Ni L2 X-ray absorption spectra
(partial fluorescence yield mode) for 100 nm NMC622|LLZO in the as-deposited state and annealed at different temperatures in air. Amplitude of
peaks normalized by the low-energy shoulder peak.
Chemical and structural characterization was done by XANES,
EXAFS, and XRD. We used samples with the same geometry for all
the characterization techniques. These characterization techniques
complemented each other and together led to a reliable identification
of secondary phases and degradation behavior. XANES and EXAFS
give information regarding oxidation state and local chemical and
coordination environment of each element. By combining data
obtained from those two techniques, and from XRD, we conclude
which secondary phases have formed.
The effect of secondary-phase formation on charge transfer at the
interface was studied by EIS. We used symmetric cells with NMC622
on both sides of the LLZO pellets. This approach let us study the
effect of the cathode|electrolyte interface only, without a Li|electrolyte
interface, and has simplified the analysis. NMC622 thin-film layers
(60 nm thick) were deposited on both sides of LLZO by RF
sputtering. Annealing conditions were the same with the samples that
we used for chemical characterization. After heat treatment, the 60 nm
thick Au layers were deposited on both sides as a current collector.
Experimental method details are provided in the Supporting
Information.
LLZO and out of the detection depth or a reaction and phase
change involving these elements.
Interestingly, the shape of feature A changed when the
sample was annealed at 700 °C, as shown in Figure 1b, where
the intensity is normalized by the magnitude of feature A. The
sample annealed at 700 °C shows a low-energy shoulder at 529
eV (A-1). This feature has not been reported by other XAS
studies on NMC, even when NMC was electrochemically
cycled.33,34 Because this feature appears after exposure to
elevated temperature, we believe that it is a consequence of a
major change in the chemical environment of the transition
metals. The meaning of the feature will be further discussed
after we present the changes in the Ni and Co L2,3-edges.
The loss of interfacial area can be characterized by
examining the O K-edge XAS features. Feature B (532−
533.5 eV) in Figure 1 originates from bulk LLZO.21 Because
feature B is clearly seen from the as-deposited sample, the
detection depth of the O K-edge XAS includes all of the
NMC622 film and the upper portion of the LLZO pellet. Lack
of feature B from the sample annealed at 700 °C indicates that
there is no LLZO left within the same detection depth. LLZO
near the NMC interface had been completely transformed to
secondary phases. Considering the geometry of the sample, the
formed secondary phases completely covered the NMC622|
LLZO interface.
Figure 2a shows the La M5 edge, La M4 edge, Ni L3 edge,
and Ni L2 edge for each sample. We normalized data by the
magnitude of La M5 edge to compare the data. The M4 and M5
edges are evident from the as-deposited sample. We attributed
the changes in the shape and magnitude of these peaks to
interfacial reactions. By comparing the magnitude of the La
and Ni edges, we can see a decrease of Ni/La elemental ratio
within the detection regime as annealing temperature
increased. This implies Ni diffusion into LLZO away from
the XAS detection depth, similar to the behavior deduced from
Figure 1. In addition, Figure 2b indicates that Ni gets more
oxidized as the annealing temperature increased. In the figure,
data for Ni L2 edge were normalized with the magnitude of
low-energy portion of the edge for comparison. Spectral weight
shift to higher energy is explained by oxidation of Ni.34−36
These findings together suggest formation of secondary phases
including Ni and La at 700 °C. Further justification will be
given by the following data.
3. RESULTS
3.1. Oxidation State Obtained by XAS. The oxidation
state and coordination of each element depends on the crystal
in which the element resides in. Therefore, depending on the
crystal structure of reactants and products, the XANES spectra
present different features before and after the NMC622|LLZO
interface reaction. By analyzing the spectra, we narrowed down
the possible set of candidates for the formed secondary phases.
We studied the near-edge regions of O K-edge, Ni L-edge, La
M-edge, Co L-edge, Ni K-edge, and Co K-edge.
Figure 1 shows the O K-edge spectra for the 100 nm
NMC622|LLZO samples, which were annealed at different
temperatures. Feature A (528−532 eV) comes from the
electronic transition stemming from hybridization of 3d
orbitals of transition metals and 2p orbital of oxygen.30,31
According to former studies on LLZO and Li2CO3 including
our former work, the change of shape and magnitude of this
feature are solely dependent on the chemical environment
change of transition metals (Ni, Co, and Mn), while O K-edge
spectra from LLZO and Li2CO3 do not have any features in
this energy range.20,21,32 As seen in Figure 1a, the magnitude of
feature A gradually decreased as annealing temperature
increased. This implies loss of transition metal atoms from
the detection depth of XAS. This could arise from either
diffusion of transition metals through NMC622|LLZO into
C
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
The Co L3 edge shifted toward higher energy, consistent
with an increase in the Co oxidation state, as the annealing
temperature increased, as shown in Figure 3. Interestingly, the
structure. However, formation of pure LaCoO3 cannot explain
the features in Co L edge shown in Figure 3. For this reason,
we propose that the secondary phase formed at 700 °C is a
doped LaCoO3. Considering the evidence of Ni/La intermix-
ing found from Figure 2, we suggest that Ni is the likely dopant
in the LaCoO3 formed here. Introducing Ni into LaCoO3 as a
B-site dopant increases the Co4+(HS) content, as reported by
Huang.38 The Co L-edge shape change shown in Huang’s data
matches well with the work of Merz et al. mentioned above,37
and with Figure 3 shown in this work. Therefore, we can
conclude that the Co oxidation state increases and becomes +4
(HS) after annealing the interface at 700 °C.
Oxidization of Ni above +2 at 700 °C, shown in Figure 2b,
requires that Ni concentration in La(Ni,Co)O3 is higher than
Co. If we denote the concentration of Ni as x in LaNixCo1−xO3
(0.5 ≤ x ≤ 1), we can rewrite the charge neutrality formula as
(La3+)(Ni2+)1−x(Ni3+)2x−1(Co4+)1−x(O2−)3 according to Per-
́
ez’s model.39 In this formula, the average oxidation state of Ni
is 4 − 1/x, and it increases from +2 to +3 as the Ni
Figure 3. Co L3, Co L2 X-ray absorption spectra (partial fluorescence
yield mode) for 100 nm NMC622|LLZO in the as-deposited state and
annealed at different temperatures in air. Amplitude of peaks
normalized by the Co L3 edge.
́
concentration increases from 0.5 to 1 in La(Ni,Co)O3. Perez’s
model also states that the oxidation state of Ni is fixed to +2
for LaNixCo1−xO3 with x ≤ 0.5. Therefore, an increase of the
oxidation state of Ni above +2 (as seen from Ni L2 edge in
Figure 2b) means x > 0.5 in LaNixCo1−xO3. Assuming that the
reaction front is at the buried interface between the LLZO and
NMC, this argument is consistent with the Ni migration out of
NMC622 (as seen from the decrease of Ni L2/La M5 ratio at
700 °C in Figure 2a), presumably occurring during La(Ni,Co)-
O3 formation.
Going back to Figure 1, oxidation of Co upon heat
treatment is seen further by analyzing the O K edge carefully.
Feature A-1 shown in Figure 1b is seen in materials in which
Co is octahedrally coordinated by an oxygen ligand with
holes.40,41 Miyoshi and Yamaguchi40 found that feature A-1
became larger for La0.6Sr0.4CoO3−δ as the oxygen partial
pressure of the environment increased. In addition, Karvonen
shape of the Co L3 edge and Co L2 edge changed drastically at
700 °C. An increase of the pre-edge feature for the Co L3 edge
at 780 eV and spectral weight shift to higher energy for the Co
L2 edge is evident. Merz et al. reported these features from
their study on doped LaCoO3 perovskites.37 The authors
compared the Co L-edge shape for LaCoO3, La0.7Ce0.3CoO3,
and La0.7Sr0.3CoO3 and found these features (the same as our
Co L2,3-edge spectrum at 700 °C) for Co4+(high spin, HS) in
La0.7Sr0.3CoO3. Notably, undoped LaCoO3 did not show
features for Co4+(HS) in their study.
In our previous work on the LiCoO2|LLZO interface,
LaCoO3 was a reaction product upon annealing.21 Therefore, it
is reasonable to assume the formation of similar perovskite
phases at the NMC622|LLZO system, as we substituted
elements of the cathode layer without changing the crystal
et al. showed that feature A-1 became larger as the oxygen
41
concentration increased in SrCoO3−δ
.
Oxygen concentration
Figure 4. Ni K-edge and (b) Co K-edge X-ray absorption spectra (fluorescence yield mode) for 100 nm NMC622|LLZO in the as-deposited state
and annealed at different temperatures in air.
D
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 5. (a) Ni K-edge and (b) Co K-edge EXAFS data in real space. (c) Single scattering paths for LiNiO2 (mp-25592) and LaNiO3 (mp-
1075921). (d) Single scattering paths for LiCoO2 (mp-24850) and LaCoO3 (mp-1068396). (Numbers after mp indicate the identity for each
crystal structure in the Materials Project Database47).
increase leads to oxidation of transition metal in those systems.
Therefore, these works give further proof to the A-1 feature,
indicating further oxidation of Co.
The near-edge regions of the Ni K-edge and Co K-edge
spectra, shown in Figure 4, provide further support for the
oxidation of Ni and Co. Data were obtained in the fluorescence
yield mode. We defined the edge by taking the energy at which
the normalized absorption value was 0.5.42 Both edges shift to
higher energy as the annealing temperature increased, showing
further oxidation of both Ni and Co.
Figure S1 shows the Mn L-edge spectra for the as-deposited
sample and for samples annealed at 300, 500, and 700 °C. All
data showed features for Mn4+ but did not show features
corresponding to Mn with other oxidation states (Mn2+ and
Mn3+).43 Therefore, the oxidation state of Mn remained the
same before and after heat treatment.
Therefore, scattering paths obtained for Ni in LaNiO3 and Co
in LaCoO3 could be used to model scattering path in
La(Ni,Co)O3. The same logic justifies using LiNiO2 and
LiCoO2 to model the scattering path in NMC 622.
The magnitudes of feature A and feature B increase up to
500 °C for both Ni K edge and Co K edge. This shows
crystallization of the sputtered NMC films because features A
and B are related to the first shell and second shell of layered
oxide structure, respectively.44−46 This crystallization trend was
also reported by our former study on annealing of LiCoO2
thin-film|LLZO system.21 However, the Ni K-edge EXAFS
deviated from this trend at 700 °C. The magnitude of feature B
dropped in comparison to the sample annealed at 500 °C,
while the magnitude of feature A remained almost the same.
Moreover, the magnitude of feature C increased compared to
other temperature conditions.
3.1.1. Local Chemical Environment Change of Elements
Due to Secondary-Phase Formation. Above we have deduced
evidence for migration of Ni and Co out of NMC622 and
formation of La(Ni,Co)O3. Thus, we expect changes in the
atomic coordination of these elements and examine the Ni K-
edge and Co K-edge EXAFS. Figure 5a,b shows the Ni K-edge
and Co K-edge EXAFS in R-space. Figure 5c,d shows ideal
scattering paths for Ni and Co in layered oxides (LiNiO2 and
LiCoO2) and in perovskites (LaNiO3 and LaCoO3). The
scattering paths are plotted for the first, second, and third
shells for layered oxides and the first and second shells for
perovskites.
Simultaneous increase of feature C and decrease of feature B
at 700 °C can be explained by formation of the La(Ni,Co)O3
phase. Figure 5a,c shows that the scattering paths from the
second shell of NMC622 mainly contributes to feature B of the
Ni K-edge EXAFS. Therefore, Ni migrating out of NMC622 to
form the perovskite phase leads to the decrease of feature B.
Moreover, we can see that the second shell of La(Ni,Co)O3
has a major contribution to feature C. Therefore, formation of
the perovskite phase increases the feature C intensity. We also
note that the first shells of both La(Ni,Co)O3 and NMC622
contribute to feature A. Therefore, decomposition of NMC622
and formation of La(Ni,Co)O3 would not change feature A,
consistent with the lack of any change in feature A of the Ni K
edge in Figure 5a.
As can be seen in Figure 5c,d, LaNiO3 and LaCoO3 show
almost identical scattering paths around Ni and Co,
respectively. This originates from the same crystal symmetry.
E
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 6. (a) 1D XRD pattern obtained by integration of 2D XRD patterns that were measured using the Co anode, as shown in Figures S2−S5,
for the 100 nm NMC622|LLZO in the as-deposited state, and annealed at different temperatures in air. (b) 1D XRD pattern [the same region
marked with dashed lines in (a)] measured using the Cu anode for the 100 nm NMC622|LLZO annealed at 700 °C in air.
The behavior at 700 °C cannot be explained by only the
crystallization of the NMC film. Feature C also has a
contribution from the third shell of NMC622 as shown in
Figure 5c. Therefore, if NMC622 is the only phase present,
feature C increase along with feature B decrease would indicate
that the crystallinity of the third shell of NMC622 increases
while the second shell loses crystallinity. This is contradictory
to increased crystallization.
In contrast to the major changes in the Ni K-edge EXAFS,
there is only a gradual increase of features A, B, and C of the
Co K-edge EXAFS with increasing annealing temperature. This
trend shown in Figure 5b can be explained by increased
crystallization of sputtered NMC622 films, without the need of
another phase to explain the data. This shows that the effect of
chemical environment change around Co due to formation of
La(Ni,Co)O3 is overwhelmed by crystallization of sputtered
NMC622 film. Therefore, we claim that Co does not
participate as much as Ni in perovskite formation. This is
consistent with our conclusion above based on the oxidation
states that Ni concentration in La(Ni,Co)O3 is higher than Co.
3.2. Structural Characterization of Secondary Phases.
XRD was done to verify the formation of La(Ni,Co)O3 and
identify other secondary phases that formed due to annealing.
In Figures S2−S5, we show 2D XRD images obtained using
Bruker GADDS. The intensities of each Debye ring were
integrated to give 1D plots as shown in Figure 6a.
Discontinuous Debye rings are present for all samples
(Figures S2−S5), and these originate from bulk LLZO grains
(ICDD: 00-063-0174). The as-deposited sample (Figure S2)
and the sample annealed at 300 °C (Figure S3) did not show
Debye rings for secondary phases. In contrast, samples
annealed at 500 °C (Figure S4) and 700 °C (Figure S5)
show arcs for Li2CO3 (ICDD: 00-022-1141) in addition to
LLZO. For the sample annealed at 700 °C (Figure S5), we also
identified arcs for La2Zr2O7 (ICDD: 01-070-5602). Arcs for
(222) plane (33.3° [Co anode]) and (400) plane (38.7° [Co
anode]) were clearly seen. In addition, we could differentiate
the Debye ring for the La2Zr2O7 (440) plane (55.9° [Co
anode]) with the nearby LLZO because it forms a continuous
ring unlike LLZO.
LaNi0.5Co0.5O3 reference spectra (ICDD: 00-054-0834),
respectively. In order to give further proofs for the perovskite
phase formation, we used PANalytical X’Pert Pro for a detailed
scan. The result is shown in Figure 6b. Detailed scan was done
within the angle range (A) shown in Figures 6a and S5 for the
sample annealed at 700 °C. The angle range was chosen so
that it includes the highest intensity peak for the perovskite
phase. In Figure 6b, the LLZO peaks show sharp doublets
because of Cu Kα1 and Cu Kα2 radiation. Li2CO3 (002) (31.8°
[Cu anode]) and La2Zr2O7 (400) (33.2° [Cu anode]) are
clearly identifiable. In addition, an additional small peak is
present just before the La2Zr2O7 (400) peak. This peak could
not be resolved using Bruker GADDS because of its limited
resolution. This peak fits with the (110) plane of
LaNi0.5Co0.5O3 reference spectra (ICDD: 00-054-0834)
(32.75° [Cu anode]), which is the strongest peak of the
perovskite structure. This provides clear evidence for the
formation of La(Ni,Co)O3. The crystallite size of La(Ni,Co)-
O3 was calculated using full width at half-maximum of XRD
peak corresponding to the (110) plane using the Scherrer
equation.48 The crystallite size of La(Ni,Co)O3 was found to
be 27 nm. However, this value does not necessarily correspond
to the reaction layer thickness, as the reaction layer can consist
of more than one crystal size of depth.
EXAFS data (Figure 5) shows the increase of short-range
order in the NMC622 film after annealing. The intensity of the
features corresponding to both the Ni−O bond (1st shell)
(Figure 5a) and Co−O bond (first shell) (Figure 5b) increased
as annealing temperature increased. However, XRD peaks
corresponding to the NMC phase could not be detected by
using the Bragg−Brentano geometry. We attribute this to the
reaction between NMC622 and LLZO, leading to only a small
amount of NMC622 left in the system. Because of this, XRD
with Bragg−Brentano geometry was not sensitive enough to
the NMC phase, even when EXAFS analysis showed increase
of short-range order in NMC. Signal from the bulk LLZO
pellet and from the formed secondary phases overwhelmed the
signals from the remaining NMC622 in the system. In order to
confirm the increased crystallinity of NMC, we performed
grazing incidence XRD using Rigaku SmartLab. Figure S6
shows grazing incidence XRD data for the sample annealed at
700 °C in air. A broad peak at 18.6° [Cu anode] corresponds
to the XRD peak from the (003) plane of NMC622 (ICDD:
00-066-0854). Formation of a broad peak for the (003) plane
We also expected Debye rings corresponding to La(Ni,Co)-
O3 from the sample annealed at 700 °C considering the
XANES and EXAFS results. We observed peaks at 27.0 and
47.4° corresponding to the (012) and (202) planes of
F
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 7. EIS data for Au|NMC622|LLZO|NMC622|Au symmetric cells. (a−d) EIS data for frequency range where the impedance is less than 1.0
kΩ cm2 are plotted. Data for (a) as-deposited sample (7.94 MHz to 794 Hz) and samples annealed at (b) 300 °C (6.31 MHz to 3.16 kHz), (c) 500
°C (6.31 MHz to 1.26 kHz), and (d) 700 °C (6.31 MHz to 1.58 kHz) in air for 4 h are shown. (e−h) EIS data for the frequency range where the
impedance is less than 6.0 kΩ cm2 are plotted. Data for (e) as-deposited sample (7.94 MHz to 63.1 Hz) and samples annealed at (f) 300 °C (6.31
MHz to 39.8 Hz), (g) 500 °C (6.31 MHz to 100 Hz), and (h) 700 °C (6.31 MHz to 158 Hz) in air for 4 h are shown.
has also been reported by previous woks on RF-deposited
LiCoO2 films.49,50 In addition, the peak strength for the (003)
plane increased as the incident angle decreased from 2 to 1°.
This further justifies that the peak originated from NMC622,
which had crystallized on top of the LLZO pellet. We could
also observe a peak which could potentially come from the
(104) plane of NMC622 (44.4° [Cu anode]). However, peak
assignment was difficult to confirm because of a nearby peak
from the LLZO (620) plane (44.0° [Cu anode]).
3.3. Effect of Degradation on Interfacial Resistance.
EIS allowed to quantify the impact of the interfacial chemical
degradation on the electrochemical charge-transfer properties.
The samples were Au|NMC622|LLZO|NMC622|Au symmet-
ric cells, with 60 nm thick NMC622 thin-film layers on both
sides of polished LLZO pellets. Au was deposited as a current
collector after samples were prepared and annealed in air.
Figure 7 shows the EIS data for samples with different
annealing conditions. In Figure 7a−d, EIS data are shown in a
narrower frequency range compared to Figure 7e−h, so that
the small arcs at higher frequencies are discernible.
of the LLZO pellet. The conductivity and relative permittivity
deduced from this arc are 2.7 × 10−1 mS/cm and ∼80,
respectively, consistent with values reported in former works
on LLZO.52,53
We also identified additional arcs at 4 kHz
(Rinterface1CPEinterface1) in all samples as can be seen in Figure
7. We attribute this feature to the NMC622|LLZO charge-
transfer resistance. The feature in the similar frequency range
was also attributed to the interface charge-transfer resistance in
our former work on LiCoO2|LLZO.21 In addition, a new arc
arises around 40 kHz (Rinterface2CPEinterface2) for the samples
annealed at 500 and 700 °C. We confirmed with XRD that the
sample annealed at 500 °C had Li2CO3, and the sample
annealed at 700 °C had Li2CO3, La(Ni,Co)O3, and La2Zr2O7
as secondary phases. Based on those findings, we attribute this
feature around 40 kHz to the impedances arising from
secondary phases.
Figure 8 shows comparison of cathode|electrolyte interfacial
resistance for each sample. As we used a symmetrically
We used two different equivalent circuit models to analyze
the EIS data. For the as-deposited sample and the sample
annealed at 300 °C, we used (Rbulk
(CPEtail)(L) circuit because there were two semicircles. On the
other hand, we used (Rbulk bulk)(Rinterface1CPEinterface1)-
Cbulk)(Rinterface1CPEinterface1)-
C
(Rinterface2CPEinterface2)(CPEtail)(L) circuit for samples annealed
at 500 and 700 °C as there were three apparent semicircles in
the plot. Serially connected inductor comes from the wiring of
the setting. This element has been used by works using LLZO
pellets and improved fitting the data at the high-frequency
range.51,52 The serially connected constant phase element
corresponds to the low-frequency tail shown in Figure 7.
Figure 8. ASR for a single cathode|electrolyte interface from the as-
deposited NMC622|LLZO and after annealing at different temper-
atures for 4 h in air.
We observed the high-frequency semicircle (RbulkCbulk
)
around 2.5 MHz in all samples. This comes from the geometry
G
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
constructed sample, the effect of interfacial resistance is twice
as large as compared to the cell with only one cathode|
electrolyte interface. In order to get interfacial resistance
through the single cathode|electrolyte interface, Figure 8 shows
the plots of Rinterface/2 for cathode|electrolyte interfacial
resistance for the as-deposited sample and the sample annealed
at 300 °C and of (Rinterface1 + Rinterface2)/2 for the samples
annealed at 500 and 700 °C.
As can be seen in Figure 8, the area-specific resistance
(ASR) increases with annealing temperature. The obtained
ASR arising from the interface or from the interface reactions,
for as-deposited, 300, and 500 °C treated samples are 1.4, 1.7,
and 2.1, respectively. Notably, ASR increases dramatically to
102 kΩ cm2 when the sample was annealed at 700 °C. This fits
well with the findings from XANES, EXAFS, and XRD that
major interfacial chemical and structural degradation takes
place at 700 °C. Formation of phases that do not conduct Li
ions, such as La2Zr2O7, La(Ni,Co)O3, and Li2CO3, is expected
to increase the ASR at the interface.21
needed for Li+/Ni2+ exchange is lower than that for Li+/Co3+
or Li+/Mn4+ exchange. The calculated formation energies for
the NiLi−LiNi, CoLi−LiCo, and MnLi−LiMn antisite pairs in
LiNi1/3Co1/3Mn1/3O2 are 0.57, 1.42, and 2.60 eV, respec-
tively.58 Moreover, Ni has a low migration barrier, ∼0.25 eV,
once it migrates into the Li site by Li+/Ni2+ exchange.59 This
value is on the same order of migration barrier of Li in layered
oxide cathodes.59−62 Solid-state reaction between two phases
in contact requires flux of elements toward the interface. As Ni
is the most mobile transition metal species in the NMC lattice,
Ni flux will be larger than Co and Mn. For this reason, if there
is an external chemical driver for reaction, we expect that Ni
will more readily migrate to the interface and participate in the
reaction.
The result suggests that increasing Ni content in the layered
oxide to obtain a high power and high capacity cell can lead to
inferior interfacial instability. Ni-rich cathode is already well
known to react with organic electrolyte and form SEI.63 Our
work is the first experimental demonstration that the same
problem exists in this all solid system, too.
EXAFS analysis for Mn coordination was not possible
because of the overlap of La L1 edge and Mn K edge, but XRD
and oxidation state analysis indicated no detectable reactivity
of Mn in our system. There were no Mn-containing secondary
phases detected using XRD. The oxidation state of Mn did not
change after annealing as seen by Mn L-edge spectra (Figure
S1). Therefore, we conclude that Mn is not participating in the
secondary-phase formation between LLZO and NMC622 in
our experimental conditions.
The third key point we discuss here is the inconsistency
between prior computational predictions of NMC|LLZO
interfacial reactions and our experimental findings. We propose
that this is because of not having considered the elevated
temperature and gas environment effects. Xiao et al. predicted
Li5CoO4, NiO, La2O3, Li6Zr2O7, and Li2MnO3 formation for
reaction between NMC and LLZO.13 Richards et al. predicted
La2Zr2O7, Li2NiO3, and La2O3 for LiNiO2|LLZO and
La2Zr2O7, Li2MnO3, and La2O3 for LiMnO2|LLZO.19 Both
works do not predict formation of La(Ni,Co)O3 phase, and we
did not find evidence to La2O3 or to Li-rich transition metal
oxides. Most importantly, the pervasive Li2CO3 formation in
our experimental findings is missing from previous theoretical
reports. The discrepancy between our experimental results and
prior calculations could arise from two key factors. First, the
computational predictions we see in the literature13,19 have not
considered the potential role of gaseous species in the
environment, in particular CO2 and H2O(g). We find
substantial Li2CO3 formation as a sink of Li, consumed by
CO2. As a result, this process is not permitting the formation of
the predicted lithium-rich phases such as Li5CoO4, Li2NiO3,
and Li2MnO3. Li2CO3 is also responsible for the formation of
delithiated phases. These are La2Zr2O7 that precipitate from
LLZO and La(Ni,Co)O3 that can form as a reaction product
when LLZO and NMC are delithiated and become more
unstable. As neither NMC622 nor LLZO contains carbon, the
reaction requires substantial carbon influx from outside the
sample to proceed. Because we annealed the sample in air,
CO2 is the carbon source. This shows the importance of the
gas environment in the interfacial degradation. Based on this,
4. DISCUSSION
The first key point we discuss is that the secondary phases
formed due to the interfacial degradation [La2Zr2O7, La-
(Ni,Co)O3, and Li2CO3] are not Li conductors. La2Zr2O7 has
poor Li conductivity, and it is a typical secondary phase found
in delithiated LLZO.54 La(Ni,Co)O3 is not expected to be a Li
conductor. There are perovskite Li conductors such as
(Li,La)TiO3, but Li conduction requires substantial A-site
vacancy fraction.55 Li2CO3 has much poorer Li conductivity
(∼10−3 mS/cm) at room temperature compared to LLZO.56
There is a large difference in interfacial resistances of the
sample that only contains Li2CO3 after exposure to 500 °C and
that of the sample that contains Li2CO3, La2Zr2O7, and
La(Ni,Co)O3 after exposure to 700 °C. The interfacial
resistance found from the former was 2.1 kΩ cm2 and that
from the latter was 102 kΩ cm2. This proves that formation of
interfacial products [La2Zr2O7 and La(Ni,Co)O3] beyond
Li2CO3 led to a substantial degradation of the charge-transfer
properties of the interface. This explains the stark increase of
interfacial resistance for the sample annealed at 700 °C.
Complete coverage of the interface with secondary phases
[Li2CO3, La2Zr2O7, and La(Ni,Co)O3 which are poor Li
conductors] led to poor Li transfer at the interface, with 2
orders of magnitude increase in the interface impedance.
The second key point is that Ni has higher tendency than
Co to participate in the formation of La(Ni,Co)O3, as deduced
from the Ni K-edge and Co K-edge EXAFS data taken from
the sample annealed at 700 °C. The features for La(Ni,Co)O3
in the Ni K-edge EXAFS spectra indicate that a substantial
amount of Ni experienced a chemical environment change. In
contrast, the Co K-edge EXAFS spectra could be fully
explained by crystallization of the NMC film. Therefore, only
a small portion of Co in the NMC622 film participated in
perovskite formation and majority of those remained in
NMC622.
Different reaction tendencies of transition metal cations can
originate in part from their mobility in NMC lattice. With this,
we mean two processes: (i) move of the transition metal to the
Li sites and (ii) the migration of the transition metal once it
resides at the Li sublattice. Migration of transition metal
cations to the nearby octahedral site of the Li sublattice makes
them mobile in the oxide.57 Ni is able to move to the Li sites
more readily than Co and Mn in NMC because the energy
we argue that controlling the gas environment, especially PCO ,
2
during the synthesis of these interfaces is important in
achieving better stability.
H
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Another reason behind the seeming discrepancy is likely the
effect of entropy, which becomes important at higher
temperatures. Both Xiao et al. and Richards et al. calculated
energies for phases which are in ground state or in equilibrium
at a certain composition and used those to find the reaction
giving largest driving force.13,19 However, as Xiao et al. noted
in their work, this approach could underestimate the effect of
configurational entropy at elevated temperatures.64
The results presented in this paper prompt us to hypothesize
that CO2 is the major contributor for interfacial degradation
between NMC622 and LLZO at elevated temperature. Li2CO3
forms by reacting with CO2 in ambient air. This reaction
reduces the Li content in NMC and in LLZO, rendering them
less stable. As a result, delithiated phases, La(Ni,Co)O3 and
La2Zr2O7, form between NMC and LLZO. This hypothesis
warrants more work to probe the NMC|LLZO interface after
subjecting it to different gas conditions, and that is the topic of
an upcoming paper of us.
Massachusetts 02139, United States; orcid.org/0000-
0002-2688-5666; Email: byildiz@mit.edu
Authors
Younggyu Kim − Laboratory for Electrochemical Interfaces,
Department of Materials Science and Engineering,
Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States; orcid.org/0000-
0002-8489-8457
Dongha Kim − Laboratory for Electrochemical Interfaces,
Department of Materials Science and Engineering,
Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States; orcid.org/0000-
0002-8072-9808
Roland Bliem − Laboratory for Electrochemical Interfaces,
Department of Nuclear Science and Engineering,
Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States; orcid.org/0000-
0002-8714-8942
Gulin Vardar − Laboratory for Electrochemical Interfaces,
̈
5. CONCLUSIONS
Department of Nuclear Science and Engineering,
Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States; orcid.org/0000-
0001-5458-9568
In this work, we revealed products of interfacial reactions at
NMC622|LLZO during annealing, starting with onset temper-
atures as low as 500 °C in air. This was possible by using thin-
film cathodes within the detection depth of near-surface-
sensitive techniques and by combining findings from
synchrotron-based XANES and EXAFS with lab-based XRD.
Major decomposition products, Li2CO3, La2Zr2O7, and
La(Ni,Co)O3, were found in the sample annealed at 700 °C.
In contrast, such phases were not present in the sample
annealed at 300 °C, and only Li2CO3 was present in the
sample annealed at 500 °C. Identification of La(Ni,Co)O3,
with substantial Ni content, was possible by careful analysis of
XANES and EXAFS data. Ni was found to be more reactive
than Co in forming this secondary phase at the LLZO
interface. Importantly, the formed secondary phases are
detrimental to the electrochemical properties. ASR at the
cathode|electrolyte interface increased by 2 orders of
magnitude after annealing up to 700 °C compared to the as-
prepared NMC622|LLZO. Based on these results, we can
conclude that using low-temperature processing is necessary to
avoid detrimental secondary-phase formation if the production
is done in air. Considering the reactants and products of the
degrading reaction, we hypothesize that CO2 is the major
contributor for interfacial degradation between NMC622 and
LLZO at elevated temperature, and we suggest using a gas
environment that do not contain CO2 to minimize interfacial
degradation.
Iradwikanari Waluyo − National Synchrotron Light Source II,
Brookhaven National Laboratory, Upton, New York 11973,
United States; orcid.org/0000-0002-4046-9722
Adrian Hunt − National Synchrotron Light Source II,
Brookhaven National Laboratory, Upton, New York 11973,
United States
Joshua T. Wright − Advanced Photon Source, Argonne
National Laboratory, Lemont, Illinois 60439, United States
John P. Katsoudas − Advanced Photon Source, Argonne
National Laboratory, Lemont, Illinois 60439, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.chemmater.0c02261
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research has been funded by U.S. Army Research Office
through Institute for Soldier Nanotechnologies (Cooperative
Agreement Number W911NF-18-2-0048). This work made
use of the Shared Experimental Facilities supported in part by
the MRSEC Program of the National Science Foundation
under award number DMR1419807. XANES data were
acquired at 23-ID-2(IOS) of the National Synchrotron Light
Source II, a U.S. Department of Energy (DOE) Office of
Science User Facility, which is operated for the DOE Office of
Science by Brookhaven National Laboratory (contract no. DE-
SC0012704). EXAFS data were obtained at 10-ID-B of the
Advanced Photon Source, a U.S. Department of Energy
(DOE) Office of Science User Facility, which is operated for
the DOE Office of Science by Argonne National Laboratory
(contract no. DE-AC02-06CH11357). We thank Yet-Ming
Chiang and Ju Li for helpful discussions. We thank Ji-Won
Jung, Seungchan Ryu, and Md Anisur Rahman for their help
during data acquisition at synchrotron experiments and Dino
Klotz, Sang-Joon Kim, and Xiahui Yao for their help with EIS
measurements.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.chemmater.0c02261.
Experimental details, Mn L-edge XANES data, 2D XRD
patterns, and grazing incidence XRD patterns (PDF)
AUTHOR INFORMATION
■
Corresponding Author
Bilge Yildiz − Laboratory for Electrochemical Interfaces,
Department of Nuclear Science and Engineering and
Department of Materials Science and Engineering,
Massachusetts Institute of Technology, Cambridge,
I
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Study on the Interfacial Stability of a Lithium Garnet-Type Oxide
Electrolyte against Cathode Materials. ACS Appl. Energy Mater. 2018,
1, 5968−5976.
(21) Vardar, G.; Bowman, W. J.; Lu, Q.; Wang, J.; Chater, R. J.;
Aguadero, A.; Seibert, R.; Terry, J.; Hunt, A.; Waluyo, I.; Fong, D. D.;
Jarry, A.; Crumlin, E. J.; Hellstrom, S. L.; Chiang, Y.-M.; Yildiz, B.
Structure, Chemistry, and Charge Transfer Resistance of the Interface
between Li7La3Zr2O12 Electrolyte and LiCoO2 Cathode. Chem. Mater.
2018, 30, 6259−6276.
(22) Chebiam, R. V.; Kannan, A. M.; Prado, F.; Manthiram, A.
Comparison of the Chemical Stability of the High Energy Density
Cathodes of Lithium-Ion Batteries. Electrochem. Commun. 2001, 3,
624−627.
REFERENCES
■
(1) Hess, S.; Wohlfahrt-Mehrens, M.; Wachtler, M. Flammability of
Li-Ion Battery Electrolytes: Flash Point and Self-Extinguishing Time
Measurements. J. Electrochem. Soc. 2015, 162, A3084−A3097.
(2) Takada, K. Progress and Prospective of Solid-State Lithium
Batteries. Acta Mater. 2013, 61, 759−770.
(3) Manthiram, A.; Yu, X.; Wang, S. Lithium Battery Chemistries
Enabled by Solid-State Electrolytes. Nat. Rev. Mater. 2017, 2, 16103.
(4) Liu, Q.; Geng, Z.; Han, C.; Fu, Y.; Li, S.; He, Y.-b.; Kang, F.; Li,
B. Challenges and Perspectives of Garnet Solid Electrolytes for All
Solid-State Lithium Batteries. J. Power Sources 2018, 389, 120−134.
(5) Thompson, T.; Yu, S.; Williams, L.; Schmidt, R. D.; Garcia-
Mendez, R.; Wolfenstine, J.; Allen, J. L.; Kioupakis, E.; Siegel, D. J.;
Sakamoto, J. Electrochemical Window of the Li-Ion Solid Electrolyte
Li7La3Zr2O12. ACS Energy Lett. 2017, 2, 462−468.
(6) Tarascon, J.-M.; Armand, M. Issues and Challenges Facing
Rechargeable Lithium Batteries. Nature 2001, 414, 359−367.
(7) Ren, Y.; Shen, Y.; Lin, Y.; Nan, C.-W. Direct Observation of
Lithium Dendrites inside Garnet-Type Lithium-Ion Solid Electrolyte.
Electrochem. Commun. 2015, 57, 27−30.
(8) Cheng, E. J.; Sharafi, A.; Sakamoto, J. Intergranular Li Metal
Propagation through Polycrystalline Li6.25Al0.25La3Zr2O12 Ceramic
Electrolyte. Electrochim. Acta 2017, 223, 85−91.
(9) Tsai, C.-L.; Roddatis, V.; Chandran, C. V.; Ma, Q.; Uhlenbruck,
S.; Bram, M.; Heitjans, P.; Guillon, O. Li7La3Zr2O12 Interface
Modification for Li Dendrite Prevention. ACS Appl. Mater. Interfaces
2016, 8, 10617−10626.
(10) Duan, H.; Yin, Y.-X.; Shi, Y.; Wang, P.-F.; Zhang, X.-D.; Yang,
C.-P.; Shi, J.-L.; Wen, R.; Guo, Y.-G.; Wan, L.-J. Dendrite-Free Li-
Metal Battery Enabled by a Thin Asymmetric Solid Electrolyte with
Engineered Layers. J. Am. Chem. Soc. 2018, 140, 82−85.
(11) Wu, B.; Wang, S.; Lochala, J.; Desrochers, D.; Liu, B.; Zhang,
W.; Yang, J.; Xiao, J. The Role of the Solid Electrolyte Interphase
Layer in Preventing Li Dendrite Growth in Solid-State Batteries.
Energy Environ. Sci. 2018, 11, 1803−1810.
(12) Han, X.; Gong, Y.; Fu, K.; He, X.; Hitz, G. T.; Dai, J.; Pearse,
A.; Liu, B.; Wang, H.; Rubloff, G.; Mo, Y.; Thangadurai, V.;
Wachsman, E. D.; Hu, L. Negating Interfacial Impedance in Garnet-
Based Solid-State Li Metal Batteries. Nat. Mater. 2017, 16, 572.
(13) Xiao, Y.; Miara, L. J.; Wang, Y.; Ceder, G. Computational
Screening of Cathode Coatings for Solid-State Batteries. Joule 2019, 3,
1252−1275.
(14) Zhu, Y.; He, X.; Mo, Y. First Principles Study on
Electrochemical and Chemical Stability of Solid Electrolyte−
Electrode Interfaces in All-Solid-State Li-Ion Batteries. J. Mater.
Chem. A 2016, 4, 3253−3266.
(23) Zheng, J.; Liu, T.; Hu, Z.; Wei, Y.; Song, X.; Ren, Y.; Wang, W.;
Rao, M.; Lin, Y.; Chen, Z.; Lu, J.; Wang, C.; Amine, K.; Pan, F.
Tuning of Thermal Stability in Layered Li(NixMnyCoz)O2. J. Am.
Chem. Soc. 2016, 138, 13326−13334.
(24) Li, W.; Song, B.; Manthiram, A. High-Voltage Positive
Electrode Materials for Lithium-Ion Batteries. Chem. Soc. Rev. 2017,
46, 3006−3059.
̈
(25) Stohr, J. NEXAFS Spectroscopy; Springer: Berlin; London, 2011.
(26) Wang, B.; Bates, J. B.; Hart, F. X.; Sales, B. C.; Zuhr, R. A.;
Robertson, J. D. Characterization of Thin-Film Rechargeable Lithium
Batteries with Lithium Cobalt Oxide Cathodes. J. Electrochem. Soc.
1996, 143, 3203.
(27) Zhu, X.; Guo, Z.; Du, G.; Zhang, P.; Liu, H. LiCoO2 Cathode
Thin Film Fabricated by RF Sputtering for Lithium Ion Micro-
batteries. Surf. Coat. Technol. 2010, 204, 1710−1714.
(28) Kim, H.-S.; Oh, Y.; Kang, K. H.; Kim, J. H.; Kim, J.; Yoon, C. S.
Characterization of Sputter-Deposited LiCoO2 Thin Film Grown on
NASICON-Type Electrolyte for Application in All-Solid-State
Rechargeable Lithium Battery. ACS Appl. Mater. Interfaces 2017, 9,
16063−16070.
(29) Bates, J. B.; Dudney, N. J.; Neudecker, B. J.; Hart, F. X.; Jun, H.
P.; Hackney, S. A. Preferred Orientation of Polycrystalline LiCoO2
Films. J. Electrochem. Soc. 2000, 147, 59−70.
(30) de Groot, F. M. F.; Grioni, M.; Fuggle, J. C.; Ghijsen, J.;
Sawatzky, G. A.; Petersen, H. Oxygen 1s x-Ray-Absorption Edges of
Transition-Metal Oxides. Phys. Rev. B: Condens. Matter Mater. Phys.
1989, 40, 5715−5723.
(31) Suntivich, J.; Hong, W. T.; Lee, Y.-L.; Rondinelli, J. M.; Yang,
W.; Goodenough, J. B.; Dabrowski, B.; Freeland, J. W.; Shao-Horn, Y.
Estimating Hybridization of Transition Metal and Oxygen States in
Perovskites from O K -Edge X-Ray Absorption Spectroscopy. J. Phys.
Chem. C 2014, 118, 1856−1863.
(32) Cheng, L.; Crumlin, E. J.; Chen, W.; Qiao, R.; Hou, H.; Franz
Lux, S.; Zorba, V.; Russo, R.; Kostecki, R.; Liu, Z.; Persson, K.; Yang,
W.; Cabana, J.; Richardson, T.; Chen, G.; Doeff, M. The Origin of
High Electrolyte−Electrode Interfacial Resistances in Lithium Cells
Containing Garnet Type Solid Electrolytes. Phys. Chem. Chem. Phys.
2014, 16, 18294−18300.
(33) Hwang, S.; Kim, S. M.; Bak, S.-M.; Kim, S. Y.; Cho, B.-W.;
Chung, K. Y.; Lee, J. Y.; Stach, E. A.; Chang, W. Using Real-Time
Electron Microscopy To Explore the Effects of Transition-Metal
Composition on the Local Thermal Stability in Charged Lix
NiyMnzCo1−y−zO2 Cathode Materials. Chem. Mater. 2015, 27,
3927−3935.
(34) Tian, C.; Nordlund, D.; Xin, H. L.; Xu, Y.; Liu, Y.; Sokaras, D.;
Lin, F.; Doeff, M. M. Depth-Dependent Redox Behavior of
LiNi0.6Mn0.2Co0.2O2. J. Electrochem. Soc. 2018, 165, A696−A704.
(35) 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.
(15) Park, K.; Yu, B.-C.; Jung, J.-W.; Li, Y.; Zhou, W.; Gao, H.; Son,
S.; Goodenough, J. B. Electrochemical Nature of the Cathode
Interface for a Solid-State Lithium-Ion Battery: Interface between
LiCoO2 and Garnet-Li7La3Zr2O12. Chem. Mater. 2016, 28, 8051−
8059.
(16) Ren, Y.; Liu, T.; Shen, Y.; Lin, Y.; Nan, C.-W. Chemical
Compatibility between Garnet-like Solid State Electrolyte
Li6.75La3Zr1.75Ta0.25O12 and Major Commercial Lithium Battery
Cathode Materials. J. Materiomics 2016, 2, 256−264.
(17) Kim, K. H.; Iriyama, Y.; Yamamoto, K.; Kumazaki, S.; Asaka,
T.; Tanabe, K.; Fisher, C. A. J.; Hirayama, T.; Murugan, R.; Ogumi, Z.
Characterization of the Interface between LiCoO2 and Li7La3Zr2O12
in an All-Solid-State Rechargeable Lithium Battery. J. Power Sources
2011, 196, 764−767.
(18) Uhlenbruck, S.; Dornseiffer, J.; Lobe, S.; Dellen, C.; Tsai, C.-L.;
Gotzen, B.; Sebold, D.; Finsterbusch, M.; Guillon, O. Cathode-
Electrolyte Material Interactions during Manufacturing of Inorganic
Solid-State Lithium Batteries. J. Electroceram. 2017, 38, 197−206.
(19) Richards, W. D.; Miara, L. J.; Wang, Y.; Kim, J. C.; Ceder, G.
Interface Stability in Solid-State Batteries. Chem. Mater. 2016, 28,
266−273.
(36) Xu, J.; Hu, E.; Nordlund, D.; Mehta, A.; Ehrlich, S. N.; Yang,
X.-Q.; Tong, W. Understanding the Degradation Mechanism of
Lithium Nickel Oxide Cathodes for Li-Ion Batteries. ACS Appl. Mater.
Interfaces 2016, 8, 31677−31683.
(20) Zhang, N.; Long, X.; Wang, Z.; Yu, P.; Han, F.; Fu, J.; Ren, G.;
Wu, Y.; Zheng, S.; Huang, W.; Wang, C.; Li, H.; Liu, X. Mechanism
J
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
̈
(37) Merz, M.; Nagel, P.; Pinta, C.; Samartsev, A.; Lohneysen, H. v.;
Wissinger, M.; Uebe, S.; Assmann, A.; Fuchs, D.; Schuppler, S. X-Ray
Absorption and Magnetic Circular Dichroism of LaCoO3,
La0.7Ce0.3CoO3, and La0.7Sr0.3CoO3 Films: Evidence for Cobalt-
Valence-Dependent Magnetism. Phys. Rev. B: Condens. Matter Mater.
Phys. 2010, 82, 174416.
(54) Chen, Y.; Rangasamy, E.; dela Cruz, C. R.; Liang, C.; An, K. A
Study of Suppressed Formation of Low-Conductivity Phases in
Doped Li7La3Zr2O12 Garnets by in Situ Neutron Diffraction. J. Mater.
Chem. A 2015, 3, 22868−22876.
́
́
(55) Alonso, J. A.; Sanz, J.; Santamaría, J.; Leon, C.; Varez, A.;
+
́
Fernandez-Díaz, M. T. On the Location of Li Cations in the Fast Li-
Cation Conductor La0.5Li0.5TiO3 Perovskite. Angew. Chem., Int. Ed.
(38) Huang, M.-J. The Electronic Structure and Magnetic
Interactions in the Mixed Transition-Metal Oxide La(Co,Ni)O3
Studied by X-Ray Absorption Spectroscopies. Doctor of Science,
2000, 39, 619−621.
(56) Ebrahiminia, M.; Hooper, J.; Bedrov, D. Structural, Mechanical,
and Dynamical Properties of Amorphous Li2CO3 from Molecular
Dynamics Simulations. Crystals 2018, 8, 473.
Karlsruher Instituts fur Technologie, 2016.
̈
́
(39) Perez, J.; García, J.; Blasco, J.; Stankiewicz, J. Spin-Glass
Behavior and Giant Magnetoresistance in the (RE)Ni0.3Co0.7O3 (RE
= La, Nd, Sm) System. Phys. Rev. Lett. 1998, 80, 2401−2404.
(40) Miyoshi, S.; Yamaguchi, S. Soft X-Ray Absorption and
Photoemission Study on the Electronic Structure of (La,Sr)CoO3.
ECS Trans. 2015, 68, 591−597.
(41) Karvonen, L.; Valkeapaa, M.; Liu, R.-S.; Chen, J.-M.; Yamauchi,
H.; Karppinen, M. O-K and Co-L XANES Study on Oxygen
Intercalation in Perovskite SrCoO3‑δ. Chem. Mater. 2010, 22, 70−76.
(42) Henderson, G. S.; de Groot, F. M. F.; Moulton, B. J. A. X-Ray
Absorption Near-Edge Structure (XANES) Spectroscopy. Rev.
Mineral. Geochem. 2014, 78, 75−138.
(43) Gilbert, B.; Frazer, B. H.; Belz, A.; Conrad, P. G.; Nealson, K.
H.; Haskel, D.; Lang, J. C.; Srajer, G.; De Stasio, G. Multiple
Scattering Calculations of Bonding and X-Ray Absorption Spectros-
copy of Manganese Oxides. J. Phys. Chem. A 2003, 107, 2839−2847.
(44) Bak, S.-M.; Nam, K.-W.; Chang, W.; Yu, X.; Hu, E.; 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.
(45) Aryal, S.; Timofeeva, E. V.; Segre, C. U. Structural Studies of
Capacity Activation and Reduced Voltage Fading in Li-Rich, Mn-Ni-
Fe Composite Oxide Cathode. J. Electrochem. Soc. 2018, 165, A71−
A78.
(57) Reed, J.; Ceder, G. Role of Electronic Structure in the
Susceptibility of Metastable Transition-Metal Oxide Structures to
Transformation. Chem. Rev. 2004, 104, 4513−4534.
(58) Hoang, K.; Johannes, M. Defect Physics and Chemistry in
Layered Mixed Transition Metal Oxide Cathode Materials:
(Ni,Co,Mn) vs (Ni,Co,Al). Chem. Mater. 2016, 28, 1325−1334.
(59) Gu, M.; Belharouak, I.; Genc, A.; Wang, Z.; Wang, D.; Amine,
K.; Gao, F.; Zhou, G.; Thevuthasan, S.; Baer, D. R.; Zhang, J.-G.;
Browning, N. D.; Liu, J.; Wang, C. Conflicting Roles of Nickel in
Controlling Cathode Performance in Lithium Ion Batteries. Nano
Lett. 2012, 12, 5186−5191.
̈
̈
(60) Moradabadi, A.; Kaghazchi, P. Mechanism of Li Intercalation/
Deintercalation into/from the Surface of LiCoO2. Phys. Chem. Chem.
Phys. 2015, 17, 22917−22922.
(61) Ning, F.; Li, S.; Xu, B.; Ouyang, C. Strain Tuned Li Diffusion in
LiCoO2 Material for Li Ion Batteries: A First Principles Study. Solid
State Ionics 2014, 263, 46−48.
(62) Ong, S. P.; Chevrier, V. L.; Hautier, G.; Jain, A.; Moore, C.;
Kim, S.; Ma, X.; Ceder, G. Voltage, Stability and Diffusion Barrier
Differences between Sodium-Ion and Lithium-Ion Intercalation
Materials. Energy Environ. Sci. 2011, 4, 3680.
(63) Manthiram, A. Electrical Energy Storage: Materials Challenges
and Prospects. MRS Bull. 2016, 41, 624−631.
(64) Xiao, Y.; Wang, Y.; Bo, S.-H.; Kim, J. C.; Miara, L. J.; Ceder, G.
Understanding Interface Stability in Solid-State Batteries. Nat. Rev.
Mater. 2020, 5, 105−126.
(46) Deb, A.; Bergmann, U.; Cramer, S. P.; Cairns, E. J. In Situ x-
Ray Absorption Spectroscopic Study of the Li[Ni1/3Co1/3Mn1/3]O2
Cathode Material. J. Appl. Phys. 2005, 97, 113523.
(47) Jain, A.; Ong, S. P.; Hautier, G.; Chen, W.; Richards, W. D.;
Dacek, S.; Cholia, S.; Gunter, D.; Skinner, D.; Ceder, G.; Persson, K.
A. Commentary: The Materials Project: A Materials Genome
Approach to Accelerating Materials Innovation. APL Mater. 2013,
1, 011002.
(48) Langford, J. I.; Wilson, A. J. C. Scherrer after Sixty Years: A
Survey and Some New Results in the Determination of Crystallite
Size. J. Appl. Crystallogr. 1978, 11, 102−113.
(49) Bouwman, P. J.; Boukamp, B. A.; Bouwmeester, H. J. M.;
Wondergem, H. J.; Notten, P. H. L. Structural Analysis of
Submicrometer LiCoO2 Films. J. Electrochem. Soc. 2001, 148,
A311−A317.
(50) Jeon, S.-W.; Lim, J.-K.; Lim, S.-H.; Lee, S.-M. As-Deposited
LiCoO2 Thin Film Cathodes Prepared by Rf Magnetron Sputtering.
Electrochim. Acta 2005, 51, 268−273.
(51) Rettenwander, D.; Welzl, A.; Cheng, L.; Fleig, J.; Musso, M.;
Suard, E.; Doeff, M. M.; Redhammer, G. J.; Amthauer, G. Synthesis
Crystal Chemistry, and Electrochemical Properties of
Li7−2xLa3Zr2−xMoxO12 (x = 0.1−0.4): Stabilization of the Cubic
Garnet Polymorph via Substitution of Zr4+ by Mo6+. Inorg. Chem.
2015, 54, 10440−10449.
(52) Wachter-Welzl, A.; Kirowitz, J.; Wagner, R.; Smetaczek, S.;
Brunauer, G. C.; Bonta, M.; Rettenwander, D.; Taibl, S.; Limbeck, A.;
Amthauer, G.; Fleig, J. The Origin of Conductivity Variations in Al-
Stabilized Li7La3Zr2O12 Ceramics. Solid State Ionics 2018, 319, 203−
208.
(53) Wachter-Welzl, A.; Wagner, R.; Rettenwander, D.; Taibl, S.;
Amthauer, G.; Fleig, J. Microelectrodes for Local Conductivity and
Degradation Measurements on Al Stabilized Li7La3Zr2O12 Garnets. J.
Electroceram. 2017, 38, 176−181.
K
https://dx.doi.org/10.1021/acs.chemmater.0c02261
Chem. Mater. XXXX, XXX, XXX−XXX