Journal of The
Electrochemical Society      
OPEN ACCESS
Wet-Chemical Synthesis of a Protective Coating
on NCM111 Cathode: The Quantified Effects of
Washing, Sintering and Coating
To cite this article: Liga Maskova et al 2024 J. Electrochem. Soc. 171 100520
 
View the article online for updates and enhancements.
You may also like
Ambient Storage Derived Surface
Contamination of NCM811 and NCM111:
Performance Implications and Mitigation
Strategies
Johannes Sicklinger, Michael Metzger,
Hans Beyer et al.
-
(De)Lithiation Mechanism of Hierarchically
Layered LiNi1/3Co1/3Mn1/3O2 Cathodes
during High-Voltage Cycling
Weibo Hua, Björn Schwarz, Michael
Knapp et al.
-
On the Origin of Reversible and
Irreversible Reactions in
LiNixCo(1x)/2Mn(1x)/2O2
Karin Kleiner, Claire A. Murray, Cristina
Grosu et al.
-
This content was downloaded from IP address 85.254.73.42 on 08/01/2025 at 11:28
Wet-Chemical Synthesis of a Protective Coating on NCM111
Cathode: The Quantified Effects of Washing, Sintering and
Coating
Liga Maskova,1,2 Reinis Ignatans,1 Arturs Viksna,3 Anatolijs Sarakovskis,1 Maris Knite,2 and
Gints Kucinskis1,z
1Institute of Solid State Physics, University of Latvia, Riga, LV-1063, Latvia
2Faculty of Natural Sciences and Technology, Riga Technical University, Riga, LV-1048, Latvia
3Department of Analytical Chemistry, Faculty of Medicine and Life Sciences, University of Latvia, Riga, LV-1004, Latvia
LiNixCoyMn1-x-yO2 (NCM) cathodes, especially with high Ni content, are widely expected to keep advancing the energy density of
Li-ion batteries. However, ensuring a good cycle life remains a key challenge. Applying inert protective coatings on the surface of
NCMs is a common route for mitigating surface-based degradation. In this study a sustainable ethanol-based wet-chemical coating
method for covering the material with Al2O3 is developed and demonstrated on NCM111. The effect of the synthesis procedure is
carefully evaluated to distinguish the benefits of the protective coating from the contributions of re-sintering and removal of surface
contaminants, all taking place during the synthesis of the coated material. We show that while the cycling stability is significantly
improved by the material regeneration alone (65% vs 79% state-of-health after 500 charge-discharge cycles at voltage range
2.7–4.3 V vs Li/Li+), the Al2O3-coated material displays further cycle life gains, maintaining 88% of initial capacity after 500
charge-discharge cycles. This work thus demonstrates both a sustainable wet-chemical coating method and the importance of
establishing a proper baseline for characterization of inert protective coatings in general. The importance of both gains further
prominence with the transition to inherently less stable higher Ni content NCMs.
© 2024 The Author(s). Published on behalf of The Electrochemical Society by IOP Publishing Limited. This is an open access
article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/
by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/
1945-7111/ad8483]
Manuscript submitted August 7, 2024; revised manuscript received September 24, 2024. Published October 17, 2024.
Supplementary material for this article is available online
Lithium-ion batteries (LIBs) are among the leading energy
storage technologies. The active anode and cathode materials used
in LIBs have a disproportionately large impact on the performance
of the battery and are thus a subject of continuous research.1 On the
forefront of the cathode development are LiNixCoyMn1-x-yO2
(NCM) layered transition metal oxide materials. To increase the
specific energy of NCM and minimize the use of cobalt, a critical
raw material, a key research focus has been on increasing the amount
of nickel–from NCM111 to the current state-of-art NCM811
(numbers denote the stoichiometric ratio between Ni, Co, and
Mn), to Ni-rich NCM materials with ever higher Ni content.
While Ni-rich materials have some of the highest energy densities
among Li-ion battery cathodes, as the amount of Ni is increased, and
Co content lowered, the thermal, structural, and cycling stabilities
deteriorate.2
The degradation processes typically plaguing NCM cathodes
during battery operation are cation mixing, oxygen evolution, side-
reactions with the electrolyte, and cathode-electrolyte interphase
(CEI) growth, with all intensifying as the Ni content increases.3
Additionally, surface impurities, such as residual lithium compounds
(RLCs) and transition metal (TM) carbonates (mainly
NiCO3·2Ni(OH)2·xH2O)
4 form on NCM cathode materials during
synthesis and storage. Besides being dead weight, the presence of
RLCs and TM carbonates is also detrimental to electrode preparation
and battery operation.3–5 Thus, strategies to minimize the amount of
lithium and transition metal species on the surface of NCMs are
needed.
Our meta-analysis shows that inert protective coatings syntheti-
cally formed on active material particles helps to prevent cycling
ageing of NCM materials, with protective coatings having an
outsized impact on the capacity retention of NCM111.3
Furthermore, the benefits of the inert protective coatings are still
retained with increasing Ni content (NCM811 and beyond),6–8 as
surface-based degradation is exacerbated with increased Ni content.
The protective coatings on cathode particles or on whole electrodes
can mitigate side-reactions with the electrolyte, hinder oxygen and
TM dissolution from the cathode material, scavenge HF formed due
to electrolyte decomposition, and at times improve electronic
(carbon-based coatings,9,10 conductive polymer coatings7,11) or ionic
(Li3PO4,
12 LiF,13 Li3VO4
14) conductivity. As a result, the cycle life
of the battery is extended. For most coatings the thickness should be
strictly limited to prevent an increase in electrochemically inactive
mass and thus loss of gravimetric capacity. The thickness of non-
conductive coatings should also be controlled to avoid an increase in
internal resistance.
Atomic layer deposition (ALD) and chemical vapor deposition
(CVD) are typically used in scientific literature and enable accurate
control of the thickness and uniformity of the coatings. However,
while suited for lab-scale development, these methods are time-
consuming, costly, and require instrument-specific knowledge and
skills for fabrication, thus their application on an industrial scale
remains challenging.
For cost and time efficiency one of the preferred synthesis
methods is wet chemical coating which usually involves mixing the
active material in a solvent, adding the coating precursor, and a
concluding sintering step to obtain the final product.6,15,16
Our work focuses on the development of a facile, sustainable,
ethanol-based wet chemical coating of NCM111 with Al2O3. While
many studies often employ water,8,15 toluene17 or other solvents,18
ethanol is a sustainable alternative that allows to forego the
challenging aqueous processing and minimizes the risk of loss of
lithium.19 The synthesis employs a surface-based reaction, forming
hydrogen bonds between the surface oxygen groups of the layered
transition metal oxide particles and hydrolyzed isopropoxide com-
pounds (hydroxides), followed by a subsequent sintering step.
Additionally, washing in ethanol has been shown to improve the
resistance to moisture and CO2 in air during subsequent storage.
19,20
Interestingly, sintering in conditions similar to what are being
used in the synthesis of protective coatings has been shown to also
improve the electrochemical performance through surface recon-
struction and removal of RLCs.21 If the active material used for
coating has not been handled properly at any point during its
delivery or processing, the improved cycle life observed due to thezE-mail: gints.kucinskis@cfi.lu.lv
Journal of The Electrochemical Society, 2024 171 100520
re-sintering and removal of RLCs could be falsely attributed to the
protective coating. Indeed, wet chemical coating studies typically
focus on the comparison between the pristine and coated material,
attributing all the stability and capacity improvements to the
developed coating.6,15,18 The magnitude of the positive effects that
washing and sintering have on the active material22,23 in this type of
processing thus remains elusive. To thoroughly evaluate the effect
that the inert protective coating has on the active material, in this
study, besides contrasting the performance of coated and uncoated
materials, we also systematically investigate the effects of the
washing and sintering steps on NCM, together and separately. In
addition to the electrochemical performance, we also analyze surface
chemistry and determine the amount of carbon to quantify the
changes in active material composition and subsequent improve-
ments in cycle life to washing, sintering and coating steps.
Experimental
Coating preparation.—Wet-chemical coating synthesis was
developed on a commercially sourced pristine NCM111 (MTI
Corp.), which had been stored in a vacuum desiccator for approxi-
mately a year. Aluminium isopropoxide (AIP) was used as raw
material for the coating. The process of the coating synthesis is
described below.
1. AIP was dissolved in anhydrous ethanol.
2. NCM111 active material was then added to the mixture so that
the weight proportion of Al2O3 coating in the final product
would be 3%. The mixture was stirred for 30 min at 400 rpm to
disperse the materials evenly.
3. A stoichiometric amount of H2O was then added to react with
AIP and form Al(OH)3.
4. The mixture was stirred for 2 h to allow Al(OH)3 coordination
around the active material particles.
5. The mixture was subsequently centrifuged and washed with
anhydrous ethanol twice to get rid of any uncoordinated
Al(OH)3 and other impurities.
6. The remaining slurry was transferred to a crucible and dried at
80 °C overnight.
7. The following day the dried powder was sintered at 500 °C in
air for 4 h (heating rate 4 °C min−1) and cooled down to room
temperature thereafter.
8. The obtained powder was stored under Ar atmosphere (O2 and
H2O <0.5 ppm) until further processing.
The reference sample was produced following steps 2.−8. (the
exact same procedure, without adding AIP and H2O as the coating-
forming agents). The washed sample was produced following steps
2.−6. (drying in vacuum at step 6), 8. and the sintered sample was
produced by following steps 7.−8. A visualization of all sample
preparation procedures is depicted in Fig. 1.
Material characterization.—The structure of unmodified (com-
mercially sourced NCM111) and modified (coated, reference,
washed, and sintered) NCM111 materials was analyzed by powder
X-ray diffraction (XRD) analysis in Rigaku MiniFlex 600 benchtop
XRD (Cu Kα1,2). Rietveld refinement was performed by using
Profex 5.2.8 software.
The XPS measurements were carried out using a ThermoFisher
ESCALAB Xi+ instrument using a monochromatic Al Kα X-ray
source. The calibration of the binding energy scale is confirmed by
examining the sputter-cleaned Au, Ag, and Cu reference samples
that place Au 4f7/2, Ag 3d5/2, and Cu 2p3/2 peaks at 83.96 eV,
368.21 eV, and 932.62 eV, respectively. The spectra were recorded
using an X-ray beam size of 650 × 100 μm with a pass energy of
20 eV and a step size of 0.1 eV. Charge neutralizer was used in the
reported experiments. The sample powders were pressed into tablets
with 5 mm diameter and stored in an Ar filled glovebox. To
minimize the formation of undesired impurities on the surface of
the analyzed samples, an inert gas-filled transfer vessel was used to
transport the samples from the glovebox to the XPS analysis
chamber without exposure to air. Prior to the XPS analysis, the
surface of the samples was gently etched in situ using Ar+ ions.
Special care was taken to avoid altering the sample chemistry due to
potential preferential sputtering.
The surface morphology was analyzed by scanning electron
microscopy (SEM, Thermo Fisher Scientific Helios 5 UX, accel-
erating voltage 10 kEv) coupled with element distribution analysis
(EDS) for powders directly after syntheses. Accelerating voltage
applied for EDS measurements was 20 keV.
Transmission electron microscope (TEM, Fei Tecnai running at
200 kV in STEM mode) coupled with EDAX EDS detector was used
to ascertain the presence of an Al-containing coating and determine
the approximate coating thickness. A lamella was prepared for the
STEM analysis from the coated NCM111 powder by depositing a
particle on top of a Si wafer, the particle was covered with thick FIB
deposited Pt layer for the surface protection. A standard lamella
preparation method was employed further (ion beam with 30 kV
acceleration voltage was used to thin the lamella to about 100 nm
thickness, this was followed by final polishing with 5 kV accelerated
ions to reduce the thickness of the amorphized surface layers).
To determine aluminum content in the coated sample, inductively
coupled plasma mass spectrometry (8900 ICP-QQQ mass spectro-
meter, Agilent Technologies, Santa Clara, CA, USA) was used.
Quantification of impurities was done using five-point calibration
graph method by diluting certified multielement standard solution.
For the analysis of carbon traces mainly in the form of surface
impurities isotope ratio mass spectrometry (IRMS, Nu Horizon,
Wrexham, United Kingdom) coupled with Elemental analyzer
(EuroVector Euro EA3000, Pavia, Italy) was used. For the quanti-
fication of carbon, a five-point calibration graph was applied using
the laboratory reference material (2% C in MnO2).
Electrode preparation.—Electrode slurries were prepared from
all samples by combining the active material with carbon black and
PVDF binder in a 75:15:10 ratio in N-methyl-2-pyrrolidone. The
slurries were mixed by magnetic stirring at 400 rpm for 2 h,
subsequent ultrasonication at 37 kHz for 30 min, and additional
stirring at 400 rpm for 30 min. The prepared slurries were manually
tape-cast on Al foil with a final wet thickness of 100 μm. The
electrode sheets were dried at 80 °C under vacuum overnight.
Finally, electrode discs (d = 1 cm) were punched from the electrode
sheet for further battery assembly.
Battery cell assembly.—Half-cells with NCM111 electrodes
were assembled in an argon-filled glove-box (O2 and H2O amount
< 0.5 ppm) using grade 316 steel coin cell casings, Whattmann GF/
B (Cytiva, Washington, DC, USA) glass fiber separators and 1 M
Figure 1. Sample preparation procedures.
Journal of The Electrochemical Society, 2024 171 100520
LiPF6 electrolyte in ethylene carbonate—diethyl carbonate (EC-
DEC) mixture with a volume ratio of 1:1 (Sigma Aldrich, battery
grade).
Electrochemical characterization.—The assembled half-cells
were tested using a Neware BTS-4000 series battery tester at a
controlled 25.0±0.1 °C temperature in a constant current (CC)
charge and discharge mode. For cycling stability measurements all
half cells were charged and discharged at 1 C (170 mA g−1) current
density for 500 charge-discharge cycles. For rate capability mea-
surements the half cells were charged and discharged in CC mode at
current densities ranging from 0.1 C to 20 C.
Results and Discussion
Material characterization.—Washing and sintering of electrode
materials are steps which are included in all wet-chemical coating
syntheses. Usually, the coating precursor is mixed with the electrode
active material in a suspension, after which the suspension is dried
and sintered. The coated material is then contrasted with the
uncoated material and all improvements to the electrochemical
performance are often attributed to the effects of coating.
Although properly coating the cathode active materials is very likely
to improve the electrochemical performance, a considerable im-
provement of the electrochemical properties may be achieved by the
washing and sintering steps alone, especially if the source material
has been improperly stored allowing surface contaminants to form,
promoting cation mixing and surface layered-to-spinel phase
transition.22,24,25 Thus, to specifically address the effect of the
standalone coating, in this study, the effects of washing and sintering
are excluded by contrasting the electrochemical performance of the
coated active material with that of the washed and sintered material
(reference sample).
Since washing, sintering, and coating modify the surface of the
active material, the electrode preparation method must be adapted
accordingly, for the effect to be most prominent. Aggressive mixing
methods such as wet ball milling or intensive high shear mixing can
crush the secondary active material particles, revealing new un-
modified surfaces from the particle bulk, thus the improvements of
material stability resulting from the surface coating may be obscured
by the degradation of the newly exposed uncoated surfaces.26–29
Furthermore, while the homogeneity of the electrode is improved,
destroying the particle integrity by ball milling has been shown to
have a negative impact on the electrochemical performance.30 Thus,
we have opted for a relatively mild mixing method of magnetic
stirring followed by sonication to retain the particle integrity and
ensure electrode homogeneity. This enables a direct comparison
between the pristine, coated and reference materials, which might
not be possible when using more vigorous and energy-intensive
mixing methods.
Examining the XRD results (Fig. 2) of the pristine, reference, and
coated NCM111, no new phases can be seen in the modified
samples, indicating that the coating is thin and/or amorphous. No
changes were observed for the only sintered and only washed
samples either. The a lattice parameter remains constant when
modifying NCM111 (~2.8618 Å) and the c lattice parameter
(14.2373 Å) shows only a very slight increase in reference and
coated materials of less than 0.001 Å. The observed lattice para-
meters are in good agreement with those reported by others.31,32
Rietveld refinement results and statistical errors are compiled in
supplementary information Fig. S1 and Tables S1–2. An increase in
Figure 2. XRD patterns of pristine, reference and coated NCM111. In the
reference (green line) and coated (purple line) samples no new phases of
Al2O3 can be observed in comparison to the pristine NCM111 (black line)
indicating that the coating is thin and possibly amorphous.
Figure 3. SEM images of pristine (a-b), reference (c-d), and coated (e-f) NCM111. No significant change in particle morphology and size is observed from the
low-magnification images (a,c,e). In the higher-magnification image of the coated sample (f) the coating cannot be clearly discerned indicating a very thin layer
of coating.
Journal of The Electrochemical Society, 2024 171 100520
c/a ratio, such as observed here (Table S1), has been linked to
improved layered structure and, thus, improved Li+ diffusivity of
NCM cathode materials.33 Additionally, since Al3+ ions have a
smaller radius than Ni2+ and Li+ ions, substitution or doping would
be seen as a decrease in lattice parameters. Since no decrease is
observed, we can conclude that Al3+ doping does not take place in
the bulk of the coated samples or is so minuscule and surface-based
that it cannot be detected by XRD. Absence of Al3+ doping is also
substantiated in several coating studies where heating even up to
800 °C does not evoke significant Al3+ diffusion into the particle
due to presence of Mn in the lattice.15,34,35
The (003)/(104) peak intensity ratios from XRD patterns are
often used as a rule-of-thumb indication of the degree of cation
mixing in the active material–values above 1.2 indicate little to no
cation mixing. These ratios of the pristine, reference and coated
samples are 1.50, 1.47, and 1.55, respectively, showing that cation
mixing is not a principal issue in the NCM111 material. Cation
mixing determined by Rietveld refinement (Table S1) shows values
of around 2% for all samples (all values match within error bounds).
Since cobalt in NCM cathodes can stabilize the lattice and mitigate
cation mixing,36,37 cation mixing and its reversal by washing,
sintering or coating38 might become more pronounced in NCMs
with higher Ni content.
Based on the SEM images (Fig. 3), washing, sintering or the
combination of both do not affect the particle size or surface
morphology of NCM111 active material. This suggests that the
amount of impurities on the surface of NCM111, which theoretically
would be removed in the reference sample, is not significant enough
to be visible at this magnification. There is also no thick coating
visible on top of the coated NCM111 particles, which is preferred, as
the desired thickness of the synthesized coating is in the nanometer
scale. Furthermore, the average primary and secondary particle size
is not affected by sample modification. The secondary and primary
particle distribution of pristine, reference, and coated NCM111 is
presented in Fig. S2.
The nanometer-scale coating is revealed by STEM-EDS analysis
(Fig. 4). STEM bright field image (Fig. 4a) shows the edge of the
secondary particle with a 10 nm thick Al-containing coating. The red
line in Fig. 4a represents the location where EDS line spectrum
collection was performed (length of line profile was ~ 27 nm, drift
corrected spectrum was collected in 27 steps resulting in roughly one
spectrum per nm). As seen in EDS line profile Fig. 4b - a drop in Ni,
Mn and Co characteristic X-ray counts is observed with a simulta-
neous increase in Al counts, indicating that the coating is formed
right on the surface of the secondary particle. Increase in Al
characteristic X-ray counts is observed in ≈ 10–20 nm interval of
the line profile, indicating that the coating is about 10 nm thick.
Figure 4c represents integrated line profile spectrum clearly showing
presence of Al, additionally strong Ga and Pt signals are observed
originating from the deposited protective layer.
XPS analysis of the uncoated, reference and coated samples
(Fig. 5) reveals changes in their surface chemistry. All samples
display characteristic Ni, Co, and Mn peaks of the NCM cathode
material (Fig. 5a). Looking closer, only the coated sample clearly
shows the band at around 73.2 eV corresponding to Al2O3/LiAlO2
39
(Fig. 5d), the intensity of which is reduced as the sample is etched,
confirming that Al is largely localized on the surface of the sample.
A coating of such chemistry is expected to form during the sintering
step of the coating procedure since the lithium from the active
material can readily migrate into the Al2O3 framework at such high
temperatures, forming LiAlO2.
40,41
The two characteristic XPS peaks in the oxygen binding energy
region (Fig. 5c) at 529.5 eV and 531.5 eV corresponding to oxygen
in the metal oxide framework of the NCM material and oxygen from
the surface impurity species (OH− and CO3
2−),42,43 respectively, are
clearly distinguishable in all samples. The metal oxide peak of the
Figure 4. TEM image of coated NCM111 (a), excerpt of EDS line spectrum at the location indicated in image a by the red line (b), cumulative EDS spectrum of
the EDS measurements across the red line indicated in image (a) (c).
Journal of The Electrochemical Society, 2024 171 100520
coated sample also includes signal from the Al2O3 coating and is
slightly more pronounced comparing to the other samples.
As carbonate impurities on the active material particle surface
degrade the battery performance by participating in CEI formation
and electrolyte deterioration, estimation of the carbon content in the
samples could give an insight into the mechanism of ageing. The
carbon XPS spectrum shows two peaks at 284.8 eV and 289.4 eV
(Fig. 5b) corresponding to adventitious carbon (hydrocarbons
adsorbed on the surface of the particles upon exposure to air) and
carbonate impurities, respectively. The adventitious carbon peak is
notably more elevated for the pristine NCM111, indicating more
adsorbed hydrocarbons on the surface of the unmodified sample.
Modifying NCM111 by washing, sintering or both notably decreases
the intensity of the adventitious carbon, and the carbonate peak is
more prominent than that of the adsorbed carbon in these samples.
The coated sample shows the lowest carbonate signal, likely due to
Al2O3/LiAlO2 covering much of the sample surface. The largest
amount of carbonates is observed on the washed samples, as the
intensity of the carbonate peak is greater than that of the adventitious
carbon. Since washing with ethanol occurs in air and ethanol is
hygroscopic, it is possible that some water and other contaminants
like CO2 are pulled into the mixture promoting carbonate formation
on sample surface. Significantly lowered carbon content is visible in
sintered, reference and coated samples.
The XPS data aligns well with the IRMS results (Fig. 6) which
were performed to determine the amount of carbon in the sample.
Both pristine and washed samples show the highest carbon content,
while sintering reduces the carbon content by approximately half as
measured by IRMS. Note that three IRMS measurements were
performed for each sample and the error bars in Fig. 6 represent the
standard deviation.
ICP-MS analysis was used to quantify the exact concentration of
Al in the samples (Fig. S3). The determined content of Al is 0.2 wt%
or 0.8 mol%, which means that approximately 10% of the aluminum
added in the form of AIP is converted to Al2O3 or LiAlO2 after the
synthesis. Interestingly, a slight Mn deficiency was also observed in
all NMC materials. After analyzing the amount of Li, Ni, Co and
Mn, we conclude that the corresponding stoichiometry is
Li1.07(Ni0.36Co0.35Mn0.29)0.93O2.
Electrochemical characterization.—All samples display similar
charge-discharge voltage profiles (Fig. 7b), while the rate perfor-
mance (Fig. 7a) suggests small but notable differences in lithium
kinetics across the samples. Washing NCM111 appears to have a
detrimental and negative effect on the discharge capacity throughout
the whole C rate range with the largest drop in capacity being
observed at high C rates (5 C, 10 C, and 20 C), indicating slower
Figure 5. XPS survey (a), Al 2p spectra (b), C 1 s spectra (c), and O 1 s spectra (d) of pristine, reference and coated NCM111. A clearly distinguishable Al 2p
peak can be observed at around 73.2 eV for the coated sample.
Figure 6. Carbon content in all samples determined by isotope ratio mass
spectrometry. A decrease of carbon content in sintered, reference and coated
samples is observed in comparison to pristine and washed samples.
Journal of The Electrochemical Society, 2024 171 100520
electrode kinetics as compared to the pristine NCM111. Sintering
has little effect on cell rate capability at any C rate. When the two
modification methods, washing and sintering (reference sample), are
combined, it appears to have a synergistic effect on the capacity at
low C rates (Figs. 7a–7b), however, the rate capability is worse
compared to the untreated material. The slower electrode kinetics of
the washed sample are also apparent in the reference sample.
Considering the variability of the results measured in coin cells,44
the measured discharge capacities of pristine, sintered, and coated
samples may fall within the error range of one another throughout
the whole C rate range. However, improved or maintained rate
capability has been reported in other coating studies employing
LiAlO2/Al2O3
17,45–47 despite Al2O3 being an isolator. The improved
kinetics has been attributed to a decrease in internal resistance
observed in impedance studies.
As can be seen from the normalized capacity fade curve
(Fig. 7c), the capacity of pristine NCM111 falls to 80% of the
initial capacity already within the first 300 cycles, and capacity
retention after 500 cycles is around 65%. NCM materials are
known to undergo rapid degradation during cycling. The main
degradation of NCM111 is localized on the surface and thus the
observed capacity fade is mostly due to CEI formation (promoted
by surface contamination) and other surface-based degradation
mechanisms, as opposed to NCM811 or more Ni-rich NCMs
where bulk degradation becomes equally or more severe as the
surface degradation.2,48
Washing the NCM111 powder with anhydrous ethanol and
subsequent drying at 80 °C in vacuum overnight decreases the
capacity retention to around 61% after 500 charge-discharge cycles.
Although washing in ethanol is reported to partly remove LiOH and
other lithium residuals from the surface,19 some contaminants such
as TM carbonates, which notably affect the cycling performance,4
may remain on the surface and their growth may be further promoted
as washing in ethanol includes steps where the cathode material is
exposed to ambient air. From the IRMS analysis it was evident that
carbon content is not reduced after washing in ethanol, which in
combination with the results from XPS (Fig. 5b) suggest that the
carbonate content might even be slightly increased. This could at
least partly explain the rapid decrease in capacity, since carbonates
on the surface of active material particles participate in CEI
formation and electrolyte degradation. Note, however, that signifi-
cant removal of lithium from NCM was not observed by the ICP-MS
for any of the samples, and initial Coulombic efficiency of all
samples were very similar: 80.6, 83.2, 81.7, 83.0 and 82.7% for
pristine, washed, sintered, reference and coated samples respec-
tively. Such results are expected, since washing in ethanol has been
shown to dissolve insignificant amounts of Li19 and sintering can re-
insert the Li from surface impurities back into the cathode
material.21
Sintering the NCM111 powder at 500 °C in air for 4 h improves
the capacity retention of the NCM111 material to 76% SoH after 500
cycles. The improvement may be attributed to the removal of surface
impurities which can be seen from the reduced amount of C in the
sintered sample as measured in the IRMS analysis. Capacity
retention could also be improved by regeneration of any surface
rock salt phase to layered phase when sintering NCM at high
temperatures.21 Combining the two modification methods (reference
sample) produces a capacity fade curve with a capacity retention of
79% - only slightly higher than the capacity retention of the sintered
sample. The capacity fade curve of the reference sample is similar to
that of the sintered NCM111 half-cell, indicating that the modifica-
tion step mostly affecting the performance of the reference sample is
sintering. The slight increase in capacity retention compared to the
sintered sample could be due to slightly lower C content in the
reference sample, however, it could also be a result of variation in
coin cell preparation.
With a proper reference established, we finally look at the coated
sample. Further improvements in capacity retention can be seen for
the coated sample (88% after 500 charge-discharge cycles),
demonstrating that a protective coating cannot be substituted by
simply washing and sintering the material. The coated material
exhibits a slightly larger c/a ratio than pristine and reference
samples, indicating an improved 2D character of the material. This
is in good agreement with the somewhat improved lithium kinetics.
Furthermore, IRMS indicates a lower carbon content on the surface
when compared to the pristine sample. The slightly increased carbon
content when compared to the reference sample may be due to the
coating locking some of the carbon impurities on the particle surface
and not allowing them to depart during the sintering step.
Nevertheless, these impurities are not directly exposed to the
electrolyte, thus, cannot participate in its degradation. The
Al2O3/LiAlO2 coating can also act as a HF scavenger further
Figure 7. Rate capability measurement (a), charge-discharge curves at 0.1 C
(b), SoH evolution during cycling at 1 C (c) for pristine, washed in ethanol,
sintered, reference, and coated samples. Charging and discharging were
carried out at CC; voltage range for all measurements was 2.7–4.3 V; 1 C =
170 mA g−1.
Journal of The Electrochemical Society, 2024 171 100520
improving the electrochemical stability of the battery cell. All of
these factors contribute to a significantly enhanced cycling stability
of the coated material.
A drastic improvement can be seen in the capacity retention (65%
to 88% SoH after 500 charge-discharge cycles) in our and many
other results when comparing the pristine and coated sample. If the
experiments are not performed carefully, however, all of the
improved cyclability can be falsely attributed to the inert protective
coating. However, we have shown here, that some improvement can
in fact be attributed to the modification of the active material
occurring during the wet-chemical synthesis procedure–especially if
the active material has been exposed to air beforehand. In our study,
we have shown that after 500 charge-discharge cycles, a pristine,
untreated NMC111 cathode material achieves 65% SoH, while the
re-sintered sample achieves 79% SoH after the same number of
cycles. A significant further cyclability improvement is achieved for
the Al2O3/LiAlO2-coated samples, achieving 88% SoH after 500 full
charge-discharge cycles at 1 C.
Conclusions
A sustainable and non-toxic method to coat NCM111 cathode
material with an Al2O3-containing coating was developed. The
results of this work demonstrate that beyond mimicking the effects
of re-sintering, the facile wet-chemical method demonstrated a
quantifiable improvement in the cycle life of the
Al2O3/LiAlO2-coated NCM111 cathode for lithium-ion batteries,
with a capacity retention of 88% after 500 charge-discharge cycles,
compared to 79% of the reference and 66% of the untreated
NCM111 sample.
By carefully analyzing each step of the treatment, we quantify
their impact on the electrochemical performance and underline the
importance of a proper reference when benchmarking the coated
samples. The results show that the capacity retention deteriorates
after washing NCM111 in absolute ethanol due to formation of
further surface carbonate species. However, subsequent sintering
reverses the decreased performance, ultimately enhancing the cycle
life compared to the pristine NCM111.
The coated NCM111, which is benchmarked against the re-
sintered sample, exhibited excellent capacity retention owing to the
maintained layered structure, lower carbon-based impurity content
and protective layer which inhibits reactions with the electrolyte and
scavenges HF formed during cell operation. Based on the promising
results presented in this work, this method provides a pathway
towards treating NCM materials to extend their cycle life. In
combination with doping and microstructural modification, this
method could be used to extend the cycle life of NCMs with higher
Ni content and successfully mitigate surface-based degradation.
Acknowledgments
The support of the M-Era.net project “Inert Coatings for
Prevention of Ageing of NMC Cathode for Lithium-Ion Batteries”
(InCoatBat), Project Reference Number 10341 is acknowledged.
Institute of Solid-State Physics, University of Latvia as the Centre of
Excellence has received funding from the European Union’s
Horizon 2020 Framework Program H2020-WIDESPREAD-01–-
2016–2017-TeamingPhase2 under grant agreement No. 739508,
project CAMART2.
ORCID
Gints Kucinskis https://orcid.org/0000-0001-9432-2358
References
1. A. Alessia, B. Alessandro, V. G. Maria, V. A. Carlos, and B. Francesca, J. Clean.
Prod., 300, 126954 (2021).
2. H. J. Noh, S. Youn, C. S. Yoon, and Y. K. Sun, J. Power Sources, 233, 121 (2013).
3. L. Britala, M. Marinaro, and G. Kucinskis, J Energy Storage, 73, 108875 (2023).
4. J. Sicklinger, M. Metzger, H. Beyer, D. Pritzl, and H. A. Gasteiger, J. Electrochem.
Soc., 166, A2322 (2019).
5. P. Teichert, G. G. Eshetu, H. Jahnke, and E. Figgemeier, Batteries, 6, 8 (2020),
https://mdpi.com/2313-0105/6/1/8/htm.
6. H. Zhang, J. Xu, and J. Zhang, Front Mater, 6, 504473 (2019).
7. X. Xiong et al., J. Solid State Electrochem., 18, 2619 (2014).
8. S. She, Y. Zhou, Z. Hong, Y. Huang, and Y. Wu, ACS Omega, 7, 24851 (2022).
9. M. Park, X. Zhang, M. Chung, G. B. Less, and A. M. Sastry, J. Power Sources, 195,
7904 (2010).
10. X. Li et al., RSC Adv., 7, 24359 (2017), https://pubs.rsc.org/en/content/articlehtml/
2017/ra/c7ra03438e.
11. B. Li et al., ChemistrySelect, 4, 6354 (2019).
12. J. Zhu et al., J. Alloys Compd., 773, 112 (2019).
13. X. Xiong, Z. Wang, X. Yin, H. Guo, and X. Li, Mater. Lett., 110, 4 (2013).
14. Y. Huang, F. M. Jin, F. J. Chen, and L. Chen, J. Power Sources, 256, 1 (2014).
15. B. Han et al., ACS Appl. Mater. Interfaces, 9, 14769 (2017).
16. G. Kaur and B. D. Gates, J. Electrochem. Soc., 169, 043504 (2022).
17. R. S. Negi and M. T. Elm, Scientific Data, 9, 1 (2022), https://nature.com/articles/
s41597-022-01217-5.
18. D. Hu et al., J. Electroanal. Chem., 880, 114910 (2021).
19. Y. Zhou, Z. Hu, Y. Huang, Y. Wu, and Z. Hong, J. Alloys Compd., 888, 161584
(2021).
20. W.-min Liu et al., Transactions of Nonferrous Metals Society of China, 28, 1626
(2018).
21. B. Huang et al., ACS Appl. Mater. Interfaces, 11, 14076 (2019).
22. X. Xiong et al., J. Power Sources, 222, 318 (2013).
23. Y. Su et al., Front Chem, 8, 573 (2020).
24. C. Busà, M. Belekoukia, and M. J. Loveridge, Electrochim. Acta, 366, 137358
(2021).
25. H. S. Liu, Z. R. Zhang, Z. L. Gong, and Y. Yang, Electrochem. Solid-State Lett., 7
(2004).
26. H. Choi, A. R. Schuer, H. Moon, M. Kuenzel, and S. Passerini, Electrochim. Acta,
430, 141047 (2022).
27. I. Hwang, C. W. Lee, J. C. Kim, and S. Yoon, Mater. Res. Bull., 47, 73 (2012).
28. J. Zhang, J. Qiao, K. Sun, and Z. Wang, Particuology, 61, 18 (2022).
29. L. Fernandez-Diaz et al., Chem. Eng. J., 464, 142469 (2023).
30. A. Ponrouch and M. R. Palacín, J. Power Sources, 196, 9682 (2011).
31. W. Hua et al., J. Electrochem. Soc., 166, A5025 (2019), https://iopscience.iop.org/
article/10.1149/2.0051903jes.
32. A. O. Kondrakov et al., J. Phys. Chem. C, 121, 3286 (2017).
33. Y. Bentaleb et al., J. Power Sources, 195, 1510 (2010).
34. B. Han et al., ACS Appl. Mater. Interfaces, 9, 41291 (2017).
35. S. Buta, D. Morgan, A. Van der Ven, M. K. Aydinol, and G. Ceder, J. Electrochem.
Soc., 146, 4335 (1999).
36. H. Zhang and J. Zhang, eTransportation, 7, 100105 (2021).
37. L. Noerochim, S. Suwarno, N. H. Idris, and H. K. Dipojono, Batteries, 7, 84 (2021),
https://mdpi.com/2313-0105/7/4/84/htm.
38. J. Zhou, X. Zhou, W. Yu, Z. Shang, and S. Xu, Electrochemical Energy Reviews, 7,
1 (2024).
39. R. S. Negi et al., Adv. Mater. Interfaces, 9, 2101428 (2022).
40. O. Touag et al., Energy Advances, 2, 701 (2023), https://pubs.rsc.org/en/content/
articlehtml/2023/ya/d3ya00061c.
41. J. L. White, F. S. Gittleson, M. Homer, and F. El Gabaly, J. Phys. Chem. C, 124,
16508 (2020).
42. R. Li et al., RSC Adv., 9, 36849 (2019), https://pubs.rsc.org/en/content/articlehtml/
2019/ra/c9ra07873h.
43. X. Ren et al., Ionics (Kiel), 26, 1617 (2020).
44. A. Smith et al., Batter Supercaps, 6, e202300080 (2023).
45. J. Cho, Y. J. Kim, and B. Park, Chem. Mater., 12, 3788 (2000).
46. R. S. Negi, S. P. Culver, A. Mazilkin, T. Brezesinski, and M. T. Elm, ACS Appl.
Mater. Interfaces, 12, 31392 (2020).
47. A. Zhou et al., J Mater Chem A Mater, 5, 24361 (2017).
48. H. H. Ryu, K. J. Park, C. S. Yoon, and Y. K. Sun, Chem. Mater., 30, 1155 (2018).
Journal of The Electrochemical Society, 2024 171 100520