Subscriber access provided by WESTERN SYDNEY U

Article
Atomic-Level Understanding of Surface Reconstruction
Based on Li[NixMnyCo1-x-y]O2 Single-Crystal Studies
Jian Zhu, Soroosh Sharifi-Asl, Juan C. Garcia, Hakim Iddir,
Jason R Croy, Reza Shahbazian-Yassar, and Guoying Chen
ACS Appl. Energy Mater., Just Accepted Manuscript • Publication Date (Web): 30 Apr 2020
Downloaded from pubs.acs.org on May 3, 2020

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 40 ACS Applied Energy Materials

1
2
3 Atomic-Level Understanding of Surface Reconstruction Based on Li[NixMnyCo1-x-y]O2
4
5
6 Single-Crystal Studies
7
8 a b c c c
Jian Zhu, Soroosh Sharifi-Asl, Juan C. Garcia, Hakim H. Iddir, Jason R. Croy, Reza
9
10
11 Shahbazian-Yassar b and Guoying Chen a,*
12
13 a
14
Energy Storage and Distributed Resources Division, Lawrence Berkeley National Laboratory,
15
16 Berkeley, California 94720, USA
17
18 b
19 Department of Mechanical and Industrial Engineering, University of Illinois at Chicago,
20
21 Chicago, Illinois 60607, USA
22
23
c Chemical Sciences and Engineering Division, Argonne National Laboratory, Lemont, Illinois
24
25
26 60439, USA
27
28
29 * Corresponding author’s email: gchen@lbl.gov
30
31
32
33
34
35 Abstract
36
37
38 The stability of cathode particle surfaces that are directly exposed to the electrolyte is one
39
40 of the most crucial and determining factors for cathode performance at high operating voltages.
41
42
43
Theory has predicted strong dependence of surface stability on chemical compositions as well as
44
45 surface facets of layered oxides, yet conflicting results on the correlations exist as most
46
47 experimental studies focus on cycled secondary particles recovered from composite electrodes.
48
49
Herein, we synthesize well-formed Li[NixMnyCo1-x-y]O2 (NMC) single-crystal samples, carefully
50
51
52 define pristine surface properties and then monitor their evolution with cycling. Atomic-resolution
53
54 scanning transmission electron microscopy (STEM) imaging and electron energy loss
55
56
57
58 1
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 2 of 40

1
2
3 spectroscopy (EELS) analysis show the formation of a surface reconstruction layer (SRL) as well
4
5
6 as an extended surface reduction layer on pristine, Li-permeable non-(001) surfaces, even before
7
8 cycling. We reveal a transition region with chemical gradient, in which the layered structure
9
10 gradually densifies and eventually transforms into the SRL on the top surface. Contrary to these
11
12
13
observations, no SRL is observed on pristine, Li-impermeable (001) surfaces, revealing the facet-
14
15 dependent nature of surface reconstructions during particle synthesis. Upon electrochemical
16
17 cycling, significant composition- and facet-dependent SRL growth is observed. The driving force
18
19
and mechanism for surface reconstruction are further discussed. The present study provides
20
21
22 insights on the origin as well as the nature of SRLs, highlighting the significance of surface
23
24 engineering in cathode material optimization.
25
26
27
28
29
30 KEYWORDS: Li-ion battery cathodes; Li[NixMnyCo1-x-y]O2 (NMC); Surface reconstruction;
31
32
33 Single crystals; Surface-face dependent; Composition-dependent
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 2
59
60 ACS Paragon Plus Environment
Page 3 of 40 ACS Applied Energy Materials

1
2
3 1. Introduction
4
5
6
7
Due to their high practical volumetric energy densities, Ni-rich layered transition-metal
8
9 (TM) oxides (TMOs) such as Li[NixMnyCo1-x-y]O2 (NMCs, x>0.5) are considered as one of the
10
11 most promising cathode materials for next-generation lithium-ion batteries (LIBs).1-5 However,
12
13 several performance issues, such as capacity fade, impedance rise, and thermal instability, become
14
15
16 significantly worse with increasing Ni content.6-8 It is known that layered-to-spinel-to-rocksalt
17
18 phase transformations take place on the surface of cycled NMCs, a phenomenon that is exacerbated
19
20 as Ni contents increase.9-10 The formation of such SRLs may block the transport of Li ions and
21
22
23 electrons, consequently leading to impedance rise as well as capacity and energy losses. The design
24
25 of better performing surfaces, therefore, is of great importance to improved cyclability of Ni-rich
26
27 NMCs, especially under high-voltage/state of charge (SOC) operating conditions.
28
29
30 Extensive effort has been dedicated to the study of surface stability against reconstruction
31
32
33 in TMOs, including NMCs, lithium nickel cobalt aluminum oxides (NCAs) and lithium- and
34
35 manganese-rich transition-metal oxides (LMRs). In most studies, the process that leads to surface
36
37 reconstruction is considered universal on all surfaces, although only certain surface facets,
38
39
40
generally those that are not parallel to {003} planes , were studied.11-15 It is only in recent years
41
42 that local anisotropic surface reconstructions have been taken into consideration. For example,
43
44 anisotropic surface reconstruction of LiNi0.4Mn0.4Co0.18Ti00.02O2 was observed along the Li
45
46
diffusion direction after only one cycle between 2.0-4.7 V.16 In a study of the impact of electrolyte
47
48
49 additives and upper cut-off voltage on LiNi0.8Mn0.1Co0.1O2 (NMC811), SRLs were found only on
50
51 surfaces not parallel to the TM-slabs (e.g., surfaces with open Li channels), regardless of the
52
53 electrolyte additives or upper cut-off voltages used.17 A two-stage rocksalt formation and growth
54
55
56 mechanism was reveal on Ni-rich NMC materials after prolonged cycling, including 1) formation
57
58 3
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 4 of 40

1
2
3 of uniform surface reconstruction layer of ~2 nm and 2) anisotropic growth of the rocksalt phase
4
5
6 into the bulk along {104} planes controlled by the solid-liquid interface or electrolyte chemistry.18
7
8 Previous studies on LMRs suggest that the pristine surfaces of primary particles terminated with
9
10 alternating TM-O-Li slabs tend to have thicker SRLs, but no apparent facet dependence was found
11
12
13
on primary particles, post-cycling.19-23 Work on pristine NCA particles reported more severe
14
15 surface reconstruction on the Li-impermeable (003) surface due to the formation of a rocksalt
16
17 phase via the interlayer mixing of Li/TM.24 All aforementioned studies were carried out on
18
19
secondary particles. As controlling primary-particle size, shape and composition uniformity in a
20
21
22 secondary particle is inherently difficult, there is a lack of overall understanding on the relationship
23
24 among anisotropic reconstruction, composition and surface facet. Furthermore, we recently
25
26 performed detailed STEM/EELS studies to evaluate the behavior of individual primary particles
27
28
29 within secondary particles.25 It was found that primary particles in the core of the secondary
30
31 particle experience less surface reconstruction, partly because they are shielded from the contact
32
33 with the electrolyte. This suggests the likelihood of misinterpretation when primary particles from
34
35
36
secondary particles are examined. By using well-formed crystal samples in our study, potential
37
38 impact from built-in microstructures is excluded and all particles can be expected to experience
39
40 similar level of charge and discharge.
41
42
In the present study, we synthesize well-formed single crystalline NMC particles with
43
44
45 dominating (001) family facets (in platelet morphology) and investigate composition- and facet-
46
47 dependent surface reconstruction before and after electrochemical cycling. LiNi0.33Mn0.33Co0.33O2
48
49 (NMC333) and LiNi0.6Mn0.2Co0.2O2 (NMC622) single crystals with the same platelet morphology
50
51
52 were used to illustrate the changes in the SRL as a function of Ni content and surface facet. SRL
53
54 formation was characterized via atomic-resolution STEM imaging, EELS, and synchrotron X-ray
55
56
57
58 4
59
60 ACS Paragon Plus Environment
Page 5 of 40 ACS Applied Energy Materials

1
2
3 absorption spectroscopy (XAS). We found SRLs present on pristine, non-(001) surfaces
4
5
6 terminated by alternating TM-O-Li layers, whereas no SRLs were found on pristine (001) surfaces
7
8 terminated by TM slabs. In addition, higher Ni content was associated with more severe surface
9
10 reconstruction of pristine particles. Cycling-induced SRL growth was broadly observed, which
11
12
13
also showed composition- and facet-dependent behavior. The mechanism for surface
14
15 reconstruction in both pristine and cycled NMCs is discussed.
16
17
18
19
20
21 2. Experimental Section
22
23
24 2.1. Synthesis
25
26
27 The pristine NMC crystal samples were prepared via a two-step process: preparation of
28
29
30 TM hydroxide precursors by the co-precipitation method followed by annealing them with lithium
31
32 hydroxide. In a typical process, stoichiometric amounts of NiSO4·6H2O, MnSO4·H2O and
33
34 CoSO4·7H2O (Sigma-Aldrich, > 99%), together with 0.5 g polyvinylpyrrolidone (PVP, with an
35
36
37 average Mw = 1,300,000), were dissolved in distilled water to obtain a TM sulfate solution A (0.8
38
39 M). NaOH and ammonia were dissolved in distilled water to prepare a solution B, at a
40
41 concentration of 2.0 M and 0.2 M, respectively. Solution A and B were added dropwise to 1 M
42
43
44
ammonia aqueous solution under stirring for 20 h. The pH of the mixed solution was maintained
45
46 at 11-12. The resulting precipitate was filtered and washed with distilled water and ethanol several
47
48 times to remove dissolved salts. The washed particles were dried in a vacuum oven at 60 oC
49
50 overnight. The as-obtained transition-metal hydroxide precursor was then mixed with LiOH.H2O
51
52
53 (with 6 mol% excess to compensate Li loss at high temperature) and annealed together for 10 h at
54
55 850 oC and 800 oC for NMC333 and NMC622, respectively.
56
57
58 5
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 6 of 40

1
2
3
4
5
6
7
2.2. Characterization
8
9
10 Powder X-ray diffraction (PXRD) was carried out using a Bruker D2 powder X-ray
11
12 diffractometer 2P (Cu KR) = 0.15418 nm, 40 kV, 30 mA). Chemical compositions of NMC samples
13
14 were determined using inductively coupled plasma-optical emission spectrometry (ICP-OES,
15
16
17
Perkin-Elmer Optima 5300 DV). Field emission scanning electron microscope (FESEM) imaging
18
19 was carried out using a JEOL JSM-7500F field emission scanning electron microscope at an
20
21 accelerating voltage of 15 kV. Hard X-ray absorption spectroscopy (hXAS) spectra of Ni, Mn, and
22
23
Co K-edges were collected in transmission mode using a (220) monochromator at the Stanford
24
25
26 Synchrotron Radiation Light source (SSRL) beamline 2-2. Pristine and cycled NMCs, recovered
27
28 from cycled electrodes, were sandwiched between two pieces of Kapton tape in an argon-filled
29
30 glovebox for the hXAS measurements. The Si (220) monochromator was detuned (~20%?) for
31
32
33 each element before the measurement to reduce the harmonics in the X-ray beam. Energy
34
35 calibration was carried out by setting the first inflection points in the spectra of Ni, Mn, Co metal
36
37 foil references to 8333 eV, 6539 eV, and 7709 eV, respectively. X-ray absorption near edge
38
39
40
structure (XANES) spectra were processed using SIXPACK software (Sam's Interface for XAS
41
42 Package developed by Sam Webb at SSRL, the Stanford Synchrotron Radiation Laboratory). Soft
43
44 XAS (sXAS) spectra were collected at SSRL on the 31-pole wiggler beamline 10-1 in a single
45
46
load at ambient temperature under ultra-high vacuum (10U@ Torr), using the total electron yield
47
48
49 (TEY) and the fluorescence yield (FY) mode detectors. A ring current of 350 mA, a 1000 l mmU
50
51 spherical grating monochromator with 40 V entrance and exit slits, a 0.2 eV energy resolution
52
53 and a 1 mm2 beam spot were used. A thin layer of pristine powder sample, or a small piece of
54
55
56 recovered electrode, was loaded onto conductive carbon tape on an aluminum sample bar in an
57
58 6
59
60 ACS Paragon Plus Environment
Page 7 of 40 ACS Applied Energy Materials

1
2
3 argon-filled glovebox for sXAS measurement. sXAS spectra data was processed using the PyMca
4
5
6 software.26
7
8
9 Aberration corrected scanning transmission electron microscopy (AC-STEM) was
10
11 performed using a JEOL JEM-ARM200CF STEM equipped with a cold field emission gun with
12
13 0.78 Å spatial resolution. Atomic resolution images were obtained by high-angle annular dark-
14
15
16 field (HAADF) detector with a 90 mrad inner-detector angle and a 22 mrad probe convergence
17
18 angle. Electron energy loss spectroscopy (EELS) was performed using a Gatan Quantum detector.
19
20 EELS results were obtained with 0.5 eV/channel dispersion and using a 5 mm detector aperture.
21
22
23 Full-width half maximum of zero loss peak was measured 1.8 eV, which determines the energy
24
25 resolution of the obtained spectra. Energy dispersive spectroscopy (EDS) results were obtained by
26
27 an Oxford X-max 100TLE windowless X-ray detector, in the range of 0-10 keV. Sample
28
29
30
preparation was done by dispersing the cathode particles in methanol and drop-casting the solution
31
32 onto a 300-mesh carbon-coated copper grid and drying in air.
33
34
35
36
37
38 2.3. Electrochemical testing
39
40
41 To prepare the NMC composite electrodes, a slurry consisting of pristine NMC crystals,
42
43
44
acetylene black conductive additive, and poly (vinylidene fluoride) (PVDF) binder in a weight
45
46 ratio of 8:1:1 in N-methylpyrrolidone (NMP) solvent was prepared. The slurry was then coated
47
48 onto a carbon-coated aluminum foil current collector (Exopack Advanced Coatings) and dried in
49
50 vacuum at 120 oC overnight. Cathode discs with an area of 1.6 cm2 and an active mass loading of
51
52
53 ~3 mg/cm2 were cut from the coated electrode sheets. The electrode discs were galvanostatically
54
55 cycled in 2032-type coin cells with lithium metal (Alfa-Aesar) counter and reference electrodes,
56
57
58 7
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 8 of 40

1
2
3 Celgard 2325 membrane separator, and a GEN 2 electrolyte containing 1.2 M LiPF6 in ethylene
4
5
6 carbonate (EC)/ethyl methyl carbonate (EMC) (3:7 weight ratio, Novolyte Technologies Inc.).
7
8 Electrochemical measurements were carried out at ambient temperature on a VMP3 multichannel
9
10 potentiostat/galvanostat, controlled by EC-lab software (BioLogic Science Instruments), at a rate
11
12
13
of C/10. Post-cycling, NMC electrodes were recovered from the cells and rinsed with dimethyl
14
15 carbonate (DMC) before the post-mortem analysis.
16
17
18
19
20
21 3. Results and discussion
22
23
24 3.1. NMC bulk structure
25
26
27 The discrete platelet-shaped NMC333 and NMC622 crystal samples were obtained by
28
29
30 using a two-step process: coprecipitation of TM sulfate precursors followed by annealing of the
31
32 resulting intermediate with lithium hydroxide. A small amount of PVP, which is a surfactant
33
34 commonly used as a surface stabilizer, a growth modifier or nanoparticle dispersant in various
35
36
37 reactions,27 was added during the coprecipitation process to influence the nucleation speed and
38
39 crystal growth. The as-synthesized NMC samples had an average platelet size of 200 nm and a
40
41
thickness of 50 nm, as shown in Figures 1a and 1b for NMC333 and NMC622, respectively.
42
43
44 This was further confirmed by STEM imaging, where the platelet morphology with sharp edges
45
46 of well-defined facets is clearly seen on both samples (Figures 1c and 1d). EDX elemental mapping
47
48 revealed uniform distribution of the TM cations and oxygen anions at the particle level (Figures
49
50
51 S1a and S1b for NMC333 and NMC622, respectively), which is in clear contrast to the
52
53 heterogeneity often seen in conventional secondary particles with built-in grain boundaries.20
54
55 Using the same TEM and selected area electron diffraction (SAED) analysis procedures reported
56
57
58 8
59
60 ACS Paragon Plus Environment
Page 9 of 40 ACS Applied Energy Materials

1
2
3 in our previous studies,28 the single crystal nature of the samples was confirmed after verifying the
4
5
6 single crystalline orientation throughout the particles.
7
8
9 Previous DFT calculations demonstrated that, in layered LiTMO2 (TM = transition metals)
10
11 such as LiCoO2, NMC333 and NMC811, the thermodynamically equilibrated particle shapes are
12
13 enclosed by three dominant families of surface facets, namely (104), (001) and (012). 29-31 While
14
15
16 the (001) plane is parallel with the Li and TM slabs in the crystal structure and is Li-ion
17
18 impermeable, both (012) and (104) facets are terminated with alternating TM-O-Li slabs and
19
20 therefore Li-permeable (Figure S2). Among them, the surface energy of the (012) and (104) facets
21
22
23 are the highest and lowest, respectively. The crystalline planes exposed on the particle surface
24
25 were index by using the shape analysis technique reported in a recent publication.32 For both
26
27 NMC333 and NMC622, more than 85% of the platelet surface is covered by the (001) facets
28
29
30
(Figure 1e), consistent with the previous results.
31
32
33 The XRD patterns collected on the as-synthesized NMC333 and NMC622 samples (Figure
34
35 1f) confirm that both samples are phase pure with the hexagonal R - E 02 structure (R m space
36
37 group). Clear peak splitting in the (006)/(012) and (018)/(110) doublets were observed, confirming
38
39
40 the well-ordered layered structure.33 The width of the splits narrows with increasing Ni content
41
42 from NMC333 to NMC622, likely due to an increase in cation mixing in the latter due to the size
43
44 similarity between Ni2+ (0.69 Å) and Li+ (0.76 Å), as reported in the previous studies.34 Rietveld
45
46
47 refinement of the XRD patterns are shown in Figure S3 and the refined parameters are listed in
48
49 Table S1. The lattice parameters for NMC333 are a = b = 2.8937 (1) Å and c = 14.3476 (3) Å,
50
51 with a c/a ratio of 4.96. These parameters are slightly decreased in NMC622, with a = b = 2.8782
52
53
(1) Å and c = 14.2481 (2) Å, and a c/a ratio of 4.95. These values are similar to what was reported
54
55
56
57
58 9
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 10 of 40

1
2
3 in the literature.35-41 The nominal composition determined by ICP analysis is also consistent with
4
5
6 the target compositions of both samples (Table S2).
7
8
9 The electrochemical performance of NMC333 and NMC622 crystal samples were
10
11 evaluated by galvanostatically charging and discharging the half-cells at a rate of C/10, using two
12
13 voltage windows of 3.0-4.3 V and 3.0-4.6 V. The evolution of the charge-discharge profile from
14
15
16 the 1st cycle to the 100th cycle are plotted at an interval of 10 cycles and shown in Figure 2a-2d.
17
18 As might be expected, capacity retention of the NMC333 was much better than that of the
19
20 NMC622, especially when both electrodes were cycled between 3.0 and 4.3 V, where the Ni-rich
21
22
23 NMC622 reaches a higher SOC. Cycling to a higher upper cutoff voltage of 4.6 V increases the
24
25 initial discharge capacity but decreases the cycling stability of both samples, as shown in Figure
26
27 2e and 2f. This trend in capacity and stability trade-off is consistent with the previous reports on
28
29
30
NMC materials.42-43
31
32
33 After 100 cycles, the cathodes were recovered from the half-cells and subjected to hXAS
34
35 analysis in order to probe changes in average TM oxidation states in the bulk of the sample. Figure
36
37 3 compares the K-edge spectra of Ni, Co and Mn collected on the pristine and cycled NMC333
38
39
40
and NMC622 electrodes. The K-edge energy position of TMs at normalized intensity of 0.5 is
41
42 often used to index the oxidation state of TMs. A higher energy shift corresponds to an increase in
43
44 TM oxidation state and a lower energy shift corresponds to TM reduction. While Co and Mn are
45
46
in +3 and +4 in both oxides, the calculated theoretical oxidation state of Ni in NMC333 and
47
48
49 NMC622 is +2 and +2.67, respectively. The measured values in the pristine samples are consistent
50
51 with these theoretical values. After cycling, either in the potential window of 3-4.3 V or 3-4.6 V,
52
53 no obvious energy shift was observed on any TM K-edge energy positions, based on the
54
55
56 normalized intensity of 0.5. This suggests negligible changes in TM chemical oxidation state in
57
58 10
59
60 ACS Paragon Plus Environment
Page 11 of 40 ACS Applied Energy Materials

1
2
3 the bulk of the samples. However, it should be noted that the energy values at the normalized
4
5
6 intensity of 0.5 does not reflect the subtle changes in chemical environment, particularly local
7
8 structural changes due to distortion around the TM. 44-45 Careful comparison of the X-ray
9
10 absorption near edge structure (XANES) before and after cycling shows certain degree of spectral
11
12
13
shape change in all three TMs, with the most significant difference found in Mn K-edge spectra.
14
15 The pre-edge peaks of Mn and Co (Figure 3 insets), on the other hand, remains similar before and
16
17 after cycling. Spectral shape change in Ni K-edge is more substantial in cycled NMC622 than that
18
19
in cycled NMC333, indicating enhanced structural distortion in the former with a higher Ni content.
20
21
22 These subtle differences in spectral features are likely caused by some Li loss and the presence of
23
24 vacant Li sites in crystal structure. At a discharge cutoff voltage of 3 V, it is conceivable that some
25
26 Li sites remain inaccessible due to impedance rise with cycling.
27
28
29
30
31
32
33 3.2. NMC surface reconstruction
34
35
36 3.2.1. Pristine NMCs
37
38
39 Layered structure with the R3m space group was further confirmed on the pristine particles
40
41 by using atomic-resolution HAADF STEM imaging. As shown in Figure 4, the TM layers and Li
42
43
44
layers are stacked along the c-axis. Cation disorder in NMC333 was negligible as bright spots were
45
46 not observed in the dark Li slab regions. A small degree of cation disorder was seen in NMC622,
47
48 consistent with the results of some cation mixing based on the Rietveld refinement (Figure S3).
49
50
While no visible SRL was found on the (001) planes in both samples, where the surface is
51
52
53 terminated by the Li-ion impermeable TM slabs (Figure 4a and 4c), a thin surface reconstruction
54
55 layer with a thickness of ~1.0 nm and ~1.5 nm was found on the non-(001) surfaces of the pristine
56
57
58 11
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 12 of 40

1
2
3 NMC333 (Figure 4a) and NMC622 (Figure 4c) particles, respectively. The nature of the surface
4
5
6 layer was further investigated by fast Fourier transform (FFT) processing of the STEM images
7
8 taken in both side view [010] zone axis (Figure S4a and S4c) and top view [012] zone axis (Figure
9
10 S4b and S4d). We note that in order to eliminate the ambiguities that arise from the fact that
11
12
13
materials with different structures can give similar two-dimensional projections, three-dimensional
14
15 structural information was obtained by imaging more than one zone axis. In both samples, it was
16
17 found that the crystal structure of the surface layer is noticeably denser than that of the bulk, as
18
19
some lithium sites were occupied by the TMs to form anti-site defects, consistent with the previous
20
21
22 reports.22 Depending on the surfaces examined, both “spinel-like” phase (Fd3m space group,
23
24 Figure S4b and S4c) and “rocksalt-type” phase (Fm3m space group, Figure S4a and S4d) were
25
26
observed on NMC333 (Figure S4a and S4b) and NMC622 (Figure S4c and S4d). The
27
28
29 transformation from the layered hexagonal structure to spinel is known to occur when ¼ of the
30
31 TM cations migrate into the Li sites, resulting a cubic symmetry that forces the Li cations into
32
33 either vacant TM octahedral or the tetrahedral sites.12 In the rocksalt structure, TM cations also
34
35
36 occupy the Li sites, but both TM and Li cations are intermixed in the same slabs. Further
37
38 investigation on such SRL formation on the pristine NMC particles was carried out by DFT
39
40 calculations and a manuscript is currently under preparation.
41
42
43 These results reveal that facet-dependent SRL formation occurs on the pristine NMC even
44
45
46 before cycling. This is likely due to a difference in thermodynamic instability of the fresh surface
47
48 during high temperature annealing treatment. It is well known that high temperature promotes Li
49
50 loss and as a precaution, most solid-state synthesis uses an excess amount of Li precursors to
51
52
53
compensate.4, 46 During the annealing process, a driving force exists for Li to leave the oxide lattice.
54
55 Li loss can be expected on the non-(001) surface with open Li diffusion channels,20 while the
56
57
58 12
59
60 ACS Paragon Plus Environment
Page 13 of 40 ACS Applied Energy Materials

1
2
3 hindered (001) surface with alternating TM-O-Li layers largely prevents such process. The vacant
4
5
6 Li sites generated on the non-(001) surface are then available for occupancy by the anti-site defects,
7
8 mostly Ni cations due to the similar size of Ni2+ and Li+, resulting in surface densification or
9
10 reconstruction of NMCs. Surface oxygen loss may also play a role in this process.
11
12
13 The SRL formation on the pristine NMC was also found to be composition-dependent. The
14
15
16 thickness of SRL on NMC333 is ~1 nm while it increased to ~1.5 nm on NMC622. The difference
17
18 is likely a result of overall better thermal stability in NMC333.39, 47 The higher Ni content in
19
20 NMC622 also increases the probability of Ni migrating into the vacant Li sites, which is consistent
21
22
23 with the results from the DFT calculations. Note that in the current study, the pristine NMC622
24
25 sample was prepared at a slightly lower temperature of 800 oC, as compared to 850 oC used in
26
27 NMC333 synthesis. This procedure was adopted to minimize cation exchange and optimize the
28
29
30
layered structure in NMC622, following the recommendation in the previous reports. 48-49 One can
31
32 expect even higher degree of surface reconstruction on NMC622 if the same synthesis temperature
33
34 was used.
35
36
37 Further understanding on pristine surface SRL was achieved by using STEM/EELS
38
39
40
analysis. Figure 5c and 5d show the EELS data collected in the non-(001) surface region of 5 nm
41
42 (at a step size of 0.5 nm) in NMC333 and NMC622, as indicated on the STEM images in Figure
43
44 5a and 5b, respectively. Changes in O K-edge along with Mn, Ni and Co L-edges were carefully
45
46
examined, with the expanded view of the peaks are also shown in Figure S5. For NMC333, while
47
48
49 Ni L-edge spectra remained nearly unchanged from the subsurface to the surface, a shift towards
50
51 lower energy was observed on both Mn and Co L-edge spectra, suggesting a decrease in oxidation
52
53 state towards the surface.50-54 O K-edge EELS is known to be associated with the transition from
54
55
56 O 1s to O 2p-TM 3d hybridized state, which is influenced by both TM oxidation states as well as
57
58 13
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 14 of 40

1
2
3 the changes in O chemical environment.16 The intensity reduction in the pre-edge peak at ~ 534
4
5
6 eV is consistent with the observed TM reduction on the surface. For NMC622, low-energy shift
7
8 was observed on all three TMs, including Ni. Figure 5e shows the L2/L3 ratio of each transition
9
10 metal as well as the energy difference between the pre- and main peaks of the O K-edge as a
11
12
13
function of distance from the surface. While L2/L3 ratio is often used to index the oxidation state
14
15 of the investigated TM,54 with a lower value corresponding to a lower oxidation state, the energy
16
17 difference between the O K-edge peaks was recently used to quantify chemical changes in
18
19
oxides.55-56 It is evident that except Ni in NMC333 that remained at ~2+, TM reduction broadly
20
21
22 occurred in the surface region of both oxides, along with a chemical gradient of increased reduction
23
24 towards surface. Based on the energy difference between the O K-edge peaks, the estimated
25
26 thickness of the reduction layer is ~2 nm and ~3 nm for NMC333 and NMC622, respectively.
27
28
29 These values are higher than the SRL thickness determined from STEM imaging, which are ~1
30
31 nm and ~1.5 nm, respectively. The results suggest that chemical changes not only occurs in the
32
33 SRL, it also extends into the subsurface region beyond the distinct layer. The extent of TM
34
35
36
reduction, however, is significantly higher in the SRL, consistent with the predominant secondary
37
38 phase in the SRL. We note that due to the relatively coarse step-size achievable in the EELS
39
40 measurements (0.5 nm per step), the thickness values are only for estimation purpose. Nonetheless,
41
42
our study not only provide particle-level confirmation on TM reduction in the SRL, it also clearly
43
44
45 reveals a transition region where the layered structure in the bulk was gradually replaced by the
46
47 densified structure in the SRL.
48
49
50 A number of studies in the literature reported the presence of SRL with reduced TMs on
51
52
53
cycled particles.12, 16, 57-58 The changes have been attributed to the effect of cycling while the
54
55 surface of pristine particles was not fully defined. We wish to emphasize that pristine samples
56
57
58 14
59
60 ACS Paragon Plus Environment
Page 15 of 40 ACS Applied Energy Materials

1
2
3 obtained from solid-state synthesis are often secondary aggregates with a variety of crystalline
4
5
6 orientations and built-in porosities and grain boundaries. As the facet- and composition-dependent
7
8 nature of SRL is evident from our study, a careful investigation on structural and compositional
9
10 inhomogeneities in the pristine sample is essential. In order to determine the real effect of cycling
11
12
13
on chemical and structural changes, one has to ensure that the same crystalline orientation is
14
15 examined for both pristine and cycled particles, as well as the same composition is ensured
16
17 throughout the particles.
18
19
20
21
22
23
3.2.2. Cycled NMCs
24
25
26
27
Post-modem analysis was carried out on the recovered cathode particles to evaluate
28
29 structural and chemical changes caused by cycling. Figure 6 shows the atomic-resolution STEM
30
31 images collected on NMC333 and NMC622 after 100 cycles between 3.0-4.6 V. It is evident that
32
33
in both samples, further SRL growth occurred during cycling, which also showed composition-
34
35
36 and facet-dependent behavior. On the non-(001) surface, the thickness of the SRL increased from
37
38 approximately 1 nm to 1.5 nm for NMC333 (Figure 6a) and 1.5 nm to 3.5 nm for NMC622 (Figure
39
40
6b), respectively. This corresponds to a net growth of 50% and 135% after 100 cycles, nearly 3x
41
42
43 faster in the higher-Ni content NMC622 sample. Although no SRL was observed on the Li-
44
45 impermeable (001) surface of pristine NMC333 or NMC622, cycling-induced surface
46
47
transformation was observed in both cases. As shown in Figure 6c and 6d, a thin SRL with ~3
48
49
50 atomic layer and ~1.5 nm were found on (001) surface of the cycled NMC333 and NMC622,
51
52 respectively. Surface densification was further evidenced by the derived intensity profiles shown
53
54 in Figure 6e and 6f, revealing the migration and occupancy of TMs into the Li layer on the surface.
55
56
57
58 15
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 16 of 40

1
2
3 Compared to the non-(001) surfaces, the growth of the SRL on the Li-impermeable (001) surface
4
5
6 is significantly less severe, confirming the facet-dependent nature of SRL growth due to cycling.
7
8 Further FFT processing of the STEM images confirms the presence of both “spinel-like” and
9
10 “rocksalt-type” structure in the densified surface layer. Factors influencing surface transformation
11
12
13
during NMC cycling were recently investigated by HAADF-STEM imaging and the results are
14
15 being summarized in a separate manuscript.
16
17
18 Cycling-induced chemical changes in the SRL were evaluated by EELS analysis of O K-
19
20 edge along with Mn, Ni and Co L-edges. Figure 7c and 7d show the spectra collected in the
21
22
23 indicated area in the STEM images from the non-(001) surface of NMC333 (Figure 7a) and
24
25 NMC622 (Figure 7b), respectively. The expanded view of the peaks are also shown in Figure S6.
26
27 Similar to the observation on the pristine samples, while Ni remained nearly unchanged at 2+ in
28
29
30
the cycled NMC333, chemical reduction in Mn and Co was broadly detected in the surface region
31
32 of both cycled samples. Figure 7e shows the L2/L3 ratio of each transition metal as well as the
33
34 energy difference between the pre- and main peaks of the O K-edge as a function of distance from
35
36 the surface. Compared to the pristine NMCs, more severe reduction as well as a thicker reduction
37
38
39 layer with larger chemical gradient were found on the cycled particles. Based on the energy
40
41 difference between the O K-edge peaks, the estimated thickness of the reduction layer with the
42
43 chemical gradient increased from approximately 2 to 3 nm and 3 to 4 nm for NMC333 and
44
45
46 NMC622, respectively. Similarly, the reduction layer is generally thicker than that of the SRL
47
48 determined by STEM imaging, which were ~1.5 and 3.5 nm, respectively. The extent of TM
49
50 reduction was significantly more severe in NMC622, confirming the role of chemical composition
51
52
53
in surface stability. These results demonstrate that cycling not only promotes further growth of
54
55 SRL, it also modifies the chemical nature of the SRL.
56
57
58 16
59
60 ACS Paragon Plus Environment
Page 17 of 40 ACS Applied Energy Materials

1
2
3 While all three TMs were found in the SRL of the cycled NMC particles, a comparison of
4
5
6 the EELS spectra collected on different regions reveals Ni segregation onto the surface. Figure 8b
7
8 and 8d show Ni and Co L-edge spectra collected on cycled NMC333 and NMC622, from the
9
10 indicated areas in the STEM images shown in Figure 8a and 8c, respectively. For NMC333, the
11
12
13
peak intensity ratio between Ni L-edge and Co L-edge is significantly higher on the top surface as
14
15 compared to that in the subsurface region below 5 nm, suggesting Ni enrichment in the SRL.
16
17 Recent work based on first-principle calculations has shown that in NMCs, nonpolar surfaces such
18
19
20 as (104) surface experiences severe Ni segregation, which can facilitate the phase transformation
21
22 from the layered to a denser crystal structure, such as spinel or rocksalt.31 Our results are consistent
23
24 with the theory work, providing clear experimental evidence on the correlation between Ni
25
26
27
migration and surface reconstruction. Surface TM segregation is largely composition-dependent.
28
29 As shown in Figure 8c and 8d, the extent of surface Ni enrichment was much reduced in cycled
30
31 NMC622. The exact role of chemical composition in surface TM segregation/enrichment is under
32
33
further investigation.
34
35
36
37
38
39
40 3.2.3. Mechanisms for NMC surface reconstruction
41
42
43 As revealed by STEM imaging and EELS analysis, the SRLs on the pristine and cycled
44
45 particles have similar nature, both showing chemical gradient as well as composition- and surface
46
47
facet-dependent behavior. Previous studies have shown higher energy barrier in the migration of
48
49
50 Mn or Co into the Li-sites.59-60 In our study, upon changing the chemical composition from
51
52 NMC333 to NMC622, Ni content was increased while both Mn and Co contents were reduced to
53
54 the same degree. The observation of more severe surface reconstruction in the latter confirms that
55
56
57
58 17
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 18 of 40

1
2
3 cation exchange and/or migration of Mn and/or Co are unlikely to occur at a meaningful level. The
4
5
6 selective surface reconstruction, therefore is likely a combined result of cation exchange (inter-
7
8 plane) and cation migration (intra-plane), particularly Li/Ni exchange and Ni migration due to the
9
10 similar size of Ni and Li cations and the low energy barrier for Ni migration.20, 61 Li/Ni mixing and
11
12
13
Ni migration processes are closely related to the Li loss from the lattice, which is facilitated either
14
15 by high temperature during the synthesis16 or high voltage during electrochemical process54. Figure
16
17 9 illustrates the proposed mechanism for SRL formation on the pristine particles (Figure 9a) and
18
19
SRL growth on cycled particles (Figure 9b).
20
21
22
23 During the annealing process, both cation-ordering (which leads to formation of layered
24
25 structure) and cation-disordering (which leads to formation of cation-mixed phase) processes take
26
27 place simultaneously along with the formation of NMC crystals.62 As the processes compete one
28
29
30
another, controlling annealing time and annealing temperature are important in order to maximize
31
32 the formation of layered structure while minimizing cation mixing in the bulk. Typically,
33
34 unfavorable spinel or rocksalt structure forms when longer reaction time or higher annealing
35
36 temperature is used.2 Even in the formed layered structure, prolonged heat exposure promotes Li
37
38
39 loss on the non-(001) surface with open Li diffusion channels,20 which is often accompanied by O
40
41 loss from the lattice. The vacant Li sites are then available to be occupied by the anti-site defects,
42
43 mostly Ni cations due to the similar size of Ni2+ and Li+. The anti-site Ni cations may migrate
44
45
46 towards the surface region due to the low energy barrier of the intra-plane transport,20, 61 resulting
47
48 in surface densification or reconstruction of the non-(001) NMC surface (Figure 9a). Depending
49
50 on the extent of Li and/or O loss, both “spinel-like” and “rocksalt-type” phases can form. On the
51
52
53
other hand, the hindered (001) surface with alternating TM-O-Li layers largely prevents such
54
55 process and therefore, SRL formation is negligible on (001) surface of NMCs.
56
57
58 18
59
60 ACS Paragon Plus Environment
Page 19 of 40 ACS Applied Energy Materials

1
2
3 During electrochemically charge and discharge process, the driving force for the inter-
4
5
6 plane cation exchange and intra-plane TM transport is the repeated Li+ extraction/insertion process.
7
8 During the charge, Li cations are removed from the Li slabs, starting from the surface to the bulk
9
10 region. Delithiated NMCs are thermodynamically less stable than the pristine counterparts, and
11
12
13
oxygen loss may also take place after extensive Li extraction at high voltages.19, 63 As a result,
14
15 inter-plane Li/Ni cation site exchange broadly occurs within the NMC particle, owing to the
16
17 relatively lower formation energy of Li/Ni exchange as compared to that of Li/Co or Li/Mn
18
19
exchange. The subsequent intra-plane migration of Li-site Ni towards the surface leads to Ni
20
21
22 enrichment and SRL growth on the non-(001) surfaces (Figure 9b). It is evident that the presence
23
24 of SRL on the pristine surface is insufficient in preventing further surface reconstruction during
25
26 electrochemical cycling in the potential window of 3.0-4.6 V.
27
28
29
30
Formation of a thin SRL was also observed on the (001) surface of cycled NMCs,
31
32 especially on NMC622. It is unclear whether surface reconstruction occurs on (001) surface due
33
34 to O loss at high voltages. It is also possible that the observed SRL is simply a result of 2D
35
36 projection of (001) with terraces and defects, or the presence of a thicker SRL formed on the non-
37
38
39 (001) surface (Figure 9b). Further studies are necessary in order to understand the formation and
40
41 nature of SRL on cycled (001) surface.
42
43
44
45
46
47 4. Conclusions
48
49
50 In this study, we synthesized single-crystalline NMC particles with well-defined surface
51
52
53 facets and utilized them to investigate the composition- and facet-dependent surface reconstruction
54
55 behavior on both pristine and cycled particles. SRL selectively formed on the non-(001) surfaces
56
57
58 19
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 20 of 40

1
2
3 whereas no SRL formation was found on (001) surface of the pristine particles. A transition region
4
5
6 with chemical gradient, in which the layered structure gradually densifies and eventually
7
8 transforms into the SRL on the top surface, was clearly revealed. Cycling induces further SRL
9
10 growth, which is also composition- and facet-dependent. We propose that SRL formation during
11
12
13
synthesis and SRL growth during electrochemical cycling share similar route causes, which are
14
15 closely related to Li permeability of the facets, inter-plane Li/Ni cation exchange and intra-plane
16
17 Ni migration. Non-(001) surfaces terminated with alternating TM-O-Li slabs and open Li diffusion
18
19
channels are prone to Li loss and the subsequent Ni migration into the vacant Li sites. The resulting
20
21
22 anti-site Ni may further transport towards the surface via intra-plane diffusion along the vacant Li
23
24 channels/slabs, leading to surface reconstruction and formation of SRL enriched with Ni. These
25
26 processes were facilitated either by the high temperature annealing during the synthesis or the high
27
28
29 potential Li extraction/insertion during the electrochemical charge-discharge process. The insights
30
31 obtained in this study provide guidance on the further design and development of better-
32
33 performing oxide cathode materials in the future.
34
35
36
37
38
39
40 Acknowledgements
41
42
43 We thank Dr. Xiangyun Song at LBNL for assisting ICP measurements, Dr. Matthew
44
45
46 Latimer and Dr. Erik Nelson at SSRL for assisting hard XAS measurements, and Dr. Dennis
47
48 Nordlund for assisting soft XAS measurements. Use of the Stanford Synchrotron Radiation
49
50 Lightsource, SLAC National Accelerator Laboratory is supported by the Office of Science, Office
51
52
53
of Basic Energy Sciences of the U.S. Department of Energy under Contract No. DE-AC02-
54
55 76SF00515. This work is supported by the Assistant Secretary for Energy Efficiency and
56
57
58 20
59
60 ACS Paragon Plus Environment
Page 21 of 40 ACS Applied Energy Materials

1
2
3 Renewable Energy, Office of Vehicle Technologies of the U.S. Department of Energy under
4
5
6 Contract No. DE-AC02-05CH11231.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 21
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 22 of 40

1
2
3 References
4
5
6 1. Radin, M. D.; Hy, S.; Sina, M.; Fang, C.; Liu, H.; Vinckeviciute, J.; Zhang, M.;
7
8
9 Whittingham, M. S.; Meng, Y. S.; Ven, A. V. d., Narrowing the Gap between Theoretical and
10
11 Practical Capacities in Li Ion Layered Oxide Cathode Materials. Adv. Energy Mater. 2017, 7 (20),
12
13 1602888.
14
15
16
2. Xu, J.; Lin, F.; Doeff, M. M.; Tong, W., A review of Ni-based layered oxides for
17
18 rechargeable Li-ion batteries. J. Mater. Chem. A 2017, 5 (3), 874-901.
19
20 3. Li, W.; Song, B.; Manthiram, A., High-voltage positive electrode materials for lithium-
21
22
ion batteries. Chem. Soc. Rev. 2017, 46 (10), 3006-3059.
23
24
25 4. Zhao, J.; Zhang, W.; Huq, A.; Misture, S. T.; Zhang, B.; Guo, S.; Wu, L.; Zhu, Y.;
26
27 Chen, Z.; Amine, K.; Pan, F.; Bai, J.; Wang, F., In Situ Probing and Synthetic Control of Cationic
28
29 Ordering in Ni-Rich Layered Oxide Cathodes. Adv. Energy Mater. 2017, 7 (3), 1601266-n/a.
30
31
32 5. Schipper, F.; Erickson, E. M.; Erk, C.; Shin, J.-Y.; Chesneau, F. F.; Aurbach, D.,
33
34 Review—Recent Advances and Remaining Challenges for Lithium Ion Battery Cathodes: I.
35
36 Nickel-Rich, LiNixCoyMnzO2. J. Electrochem. Soc. 2017, 164 (1), A6220-A6228.
37
38
39
6. Jung, S.-K.; Gwon, H.; Hong, J.; Park, K.-Y.; Seo, D.-H.; Kim, H.; Hyun, J.; Yang,
40
41 W.; Kang, K., Understanding the Degradation Mechanisms of LiNi0.5Co0.2Mn0.3O2 Cathode
42
43 Material in Lithium Ion Batteries. Adv. Energy Mater. 2014, 4 (1), 1300787-n/a.
44
45
7. Manthiram, A.; Song, B.; Li, W., A perspective on nickel-rich layered oxide cathodes for
46
47
48 lithium-ion batteries. Energy Storage Materials 2017, 6, 125-139.
49
50 8. Myung, S.-T.; Maglia, F.; Park, K.-J.; Yoon, C. S.; Lamp, P.; Kim, S.-J.; Sun, Y.-K.,
51
52 Nickel-Rich Layered Cathode Materials for Automotive Lithium-Ion Batteries: Achievements and
53
54
55 Perspectives. ACS Energy Lett. 2017, 2 (1), 196-223.
56
57
58 22
59
60 ACS Paragon Plus Environment
Page 23 of 40 ACS Applied Energy Materials

1
2
3 9. Zhao, W.; Zheng, J.; Zou, L.; Jia, H.; Liu, B.; Wang, H.; Engelhard, M. H.; Wang, C.;
4
5
6 Xu, W.; Yang, Y.; Zhang, J.-G., High Voltage Operation of Ni-Rich NMC Cathodes Enabled by
7
8 Stable Electrode/Electrolyte Interphases. Adv. Energy Mater. 2018, 8 (19), 1800297.
9
10 10. Shkrob, I. A.; Gilbert, J. A.; Phillips, P. J.; Klie, R.; Haasch, R. T.; Bareno, J.; Abraham,
11
12
13
D. P., Chemical Weathering of Layered Ni-Rich Oxide Electrode Materials: Evidence for Cation
14
15 Exchange. J. Electrochem. Soc. 2017, 164 (7), A1489-A1498.
16
17 11. Zhang, X.; Belharouak, I.; Li, L.; Lei, Y.; Elam, J. W.; Nie, A.; Chen, X.; Yassar, R.
18
19
S.; Axelbaum, R. L., Structural and Electrochemical Study of Al2O3 and TiO2 Coated
20
21
22 Li1.2Ni0.13Mn0.54Co0.13O2 Cathode Material Using ALD. Adv. Energy Mater. 2013, 3 (10),
23
24 1299-1307.
25
26 12. Kim, N. Y.; Yim, T.; Song, J. H.; Yu, J.-S.; Lee, Z., Microstructural study on degradation
27
28
29 mechanism of layered LiNi0.6Co0.2Mn0.2O2 cathode materials by analytical transmission
30
31 electron microscopy. J. Power Sources 2016, 307, 641-648.
32
33 13. Zhang, C.; Liu, M.; Pan, G.; Liu, S.; Liu, D.; Chen, C.; Su, J.; Huang, T.; Yu, A.,
34
35
36
Enhanced Electrochemical Performance of LiNi0.8Co0.1Mn0.1O2 Cathode for Lithium-Ion
37
38 Batteries by Precursor Preoxidation. ACS Applied Energy Materials 2018, 1 (8), 4374-4384.
39
40 14. Ryu, H.-H.; Park, K.-J.; Yoon, C. S.; Sun, Y.-K., Capacity Fading of Ni-Rich
41
42
Li[NixCoyMn1–x–y]O2 (0.6 ` x ` 0.95) Cathodes for High-Energy-Density Lithium-Ion Batteries:
43
44
45 Bulk or Surface Degradation? Chem. Mater. 2018, 30 (3), 1155-1163.
46
47 15. Kim, J.; Cho, H.; Jeong, H. Y.; Ma, H.; Lee, J.; Hwang, J.; Park, M.; Cho, J., Self-
48
49 Induced Concentration Gradient in Nickel-Rich Cathodes by Sacrificial Polymeric Bead Clusters
50
51
52 for High-Energy Lithium-Ion Batteries. Adv. Energy Mater. 2017, 7 (12), 1602559.
53
54
55
56
57
58 23
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 24 of 40

1
2
3 16. Lin, F.; Markus, I. M.; Nordlund, D.; Weng, T.-C.; Asta, M. D.; Xin, H. L.; Doeff, M.
4
5
6 M., Surface reconstruction and chemical evolution of stoichiometric layered cathode materials for
7
8 lithium-ion batteries. Nat. Commun. 2014, 5, 3529.
9
10 17. Li, J.; Liu, H.; Xia, J.; Cameron, A. R.; Nie, M.; Botton, G. A.; Dahn, J. R., The Impact
11
12
13
of Electrolyte Additives and Upper Cut-off Voltage on the Formation of a Rocksalt Surface Layer
14
15 in LiNi0.8Mn0.1Co0.1O2 Electrodes. J. Electrochem. Soc. 2017, 164 (4), A655-A665.
16
17 18. Zou, L.; Liu, Z.; Zhao, W.; Jia, H.; Zheng, J.; Yang, Y.; Wang, G.; Zhang, J.-G.; Wang,
18
19
C., Solid–Liquid Interfacial Reaction Trigged Propagation of Phase Transition from Surface into
20
21
22 Bulk Lattice of Ni-Rich Layered Cathode. Chem. Mater. 2018, 30 (20), 7016-7026.
23
24 19. Liu, H.; Harris, K. J.; Jiang, M.; Wu, Y.; Goward, G. R.; Botton, G. A., Unraveling the
25
26 Rapid Performance Decay of Layered High-Energy Cathodes: From Nanoscale Degradation to
27
28
29 Drastic Bulk Evolution. ACS Nano 2018, 12 (3), 2708-2718.
30
31 20. Gu, M.; Belharouak, I.; Genc, A.; Wang, Z. G.; Wang, D. P.; Amine, K.; Gao, F.; Zhou,
32
33 G. W.; Thevuthasan, S.; Baer, D. R.; Zhang, J. G.; Browning, N. D.; Liu, J.; Wang, C. M.,
34
35
36
Conflicting Roles of Nickel in Controlling Cathode Performance in Lithium Ion Batteries. Nano
37
38 Lett. 2012, 12 (10), 5186-5191.
39
40 21. Dixit, H.; Zhou, W.; Idrobo, J.-C.; Nanda, J.; Cooper, V. R., Facet-Dependent Disorder
41
42
in Pristine High-Voltage Lithium–Manganese-Rich Cathode Material. ACS Nano 2014, 8 (12),
43
44
45 12710-12716.
46
47 22. Yan, P. F.; Zheng, J. M.; Zheng, J. X.; Wang, Z. G.; Teng, G. F.; Kuppan, S.; Xiao, J.;
48
49 Chen, G. Y.; Pan, F.; Zhang, J. G.; Wang, C. M., Ni and Co Segregations on Selective Surface
50
51
52 Facets and Rational Design of Layered Lithium Transition-Metal Oxide Cathodes. Adv. Energy
53
54 Mater. 2016, 6 (9), 1502455.
55
56
57
58 24
59
60 ACS Paragon Plus Environment
Page 25 of 40 ACS Applied Energy Materials

1
2
3 23. Jarvis, K.; Wang, C. C.; Varela, M.; Unocic, R. R.; Manthiram, A.; Ferreira, P. J., Surface
4
5
6 Reconstruction in Li-Rich Layered Oxides of Li-Ion Batteries. Chem. Mater. 2017, 29 (18), 7668-
7
8 7674.
9
10 24. Zhang, H.; May, B. M.; Serrano-Sevillano, J.; Casas-Cabanas, M.; Cabana, J.; Wang,
11
12
13
C.; Zhou, G., Facet-Dependent Rock-Salt Reconstruction on the Surface of Layered Oxide
14
15 Cathodes. Chem. Mater. 2018, 30 (3), 692-699.
16
17 25. Zou, L.; Zhao, W.; Jia, H.; Zheng, J.; Li, L.; Abraham, D. P.; Chen, G.; Croy, J. R.;
18
19
Zhang, J.-G.; Wang, C., The Role of Secondary Particle Structures in Surface Phase Transitions
20
21
22 of Ni-Rich Cathodes. Chem. Mater. 2020, DOI: 10.1021/acs.chemmater.9b04938.
23
24 26. Solé, V. A.; Papillon, E.; Cotte, M.; Walter, P.; Susini, J., A multiplatform code for the
25
26 analysis of energy-dispersive X-ray fluorescence spectra. Spectrochimica Acta Part B: Atomic
27
28
29 Spectroscopy 2007, 62 (1), 63-68.
30
31 27. Koczkur, K. M.; Mourdikoudis, S.; Polavarapu, L.; Skrabalak, S. E., Polyvinylpyrrolidone
32
33 (PVP) in nanoparticle synthesis. Dalton Trans. 2015, 44 (41), 17883-17905.
34
35
36
28. Chen, G. Y.; Hai, B.; Shukla, A. K.; Duncan, H., Impact of Initial Li Content on Kinetics
37
38 and Stabilities of Layered Li1+x(Ni0.33Mn0.33Co0.33)(1-x)O-2. J. Electrochem. Soc. 2012, 159
39
40 (9), A1543-A1550.
41
42
29. Garcia, J. C.; Bareno, J.; Yan, J. H.; Chen, G. Y.; Hauser, A.; Croy, J. R.; Iddir, H.,
43
44
45 Surface Structure, Morphology, and Stability of Li(Ni1/3Mn1/3Co1/3)O-2 Cathode Material. J.
46
47 Phys. Chem. C 2017, 121 (15), 8290-8299.
48
49 30. Kramer, D.; Ceder, G., Tailoring the Morphology of LiCoO2: A First Principles Study.
50
51
52 Chem. Mater. 2009, 21 (16), 3799-3809.
53
54
55
56
57
58 25
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 26 of 40

1
2
3 31. Liang, C.; Longo, R. C.; Kong, F.; Zhang, C.; Nie, Y.; Zheng, Y.; Cho, K., Ab initio
4
5
6 Study on Surface Segregation and Anisotropy of Ni-rich LiNi1-2yCoyMnyO2 (NCM) (y ` 0.1)
7
8 Cathodes. ACS Appl. Mater. Interfaces 2018, 10 (7), 6673-6680.
9
10 32. Zhu, J.; Chen, G., Single-crystal based studies for correlating the properties and high-
11
12
13
voltage performance of + ,- ). U)U /01 cathodes. J. Mater. Chem. A 2019, 7 (10), 5463-
14
15 5474.
16
17 33. Wang, F.; Xiao, S.; Chang, Z.; Yang, Y.; Wu, Y., Nanoporous LiNi1/3Co1/3Mn1/3O2
18
19
as an ultra-fast charge cathode material for aqueous rechargeable lithium batteries. Chem. Commun.
20
21
22 2013, 49 (80), 9209-9211.
23
24 34. Huang, Z.-D.; Liu, X.-M.; Oh, S.-W.; Zhang, B.; Ma, P.-C.; Kim, J.-K., Microscopically
25
26 porous, interconnected single crystal LiNi1/3Co1/3Mn1/3O2 cathode material for Lithium ion
27
28
29 batteries. J. Mater. Chem. 2011, 21 (29), 10777-10784.
30
31 35. Choi, J.; Manthiram, A., Crystal chemistry and electrochemical characterization of layered
32
33 + - 5 <U 5 <U . 1 01 and + 5 <U . 5 <U - 1 01 25`1 ` 3 cathodes. J. Power
34
35
36
Sources 2006, 162 (1), 667-672.
37
38 36. Lee, K.-S.; Myung, S.-T.; Amine, K.; Yashiro, H.; Sun, Y.-K., Structural and
39
40 Electrochemical Properties of Layered + b,b- bUb1) Co x Mn x b/b01 b2b)bOb5 b–b5 9b3b Positive
41
42
Electrode Materials for Li-Ion Batteries. J. Electrochem. Soc. 2007, 154 (10), A971-A977.
43
44
45 37. Xiao, J.; Chernova, N. A.; Whittingham, M. S., Influence of Manganese Content on the
46
47 Performance of + - 5 @U . 5 01 (0.45 ` y ` 0.60) as a Cathode Material for Li-Ion
48
49 Batteries. Chem. Mater. 2010, 22 (3), 1180-1185.
50
51
52
53
54
55
56
57
58 26
59
60 ACS Paragon Plus Environment
Page 27 of 40 ACS Applied Energy Materials

1
2
3 38. Min, K.; Kim, K.; Jung, C.; Seo, S.-W.; Song, Y. Y.; Lee, H. S.; Shin, J.; Cho, E., A
4
5
6 comparative study of structural changes in lithium nickel cobalt manganese oxide as a function of
7
8 Ni content during delithiation process. J. Power Sources 2016, 315, 111-119.
9
10 39. Sun, H.; Zhao, K., Electronic Structure and Comparative Properties of LiNixMnyCozO2
11
12
13
Cathode Materials. J. Phys. Chem. C 2017, 121 (11), 6002-6010.
14
15 40. Sun, H.-H.; Manthiram, A., Impact of Microcrack Generation and Surface Degradation on
16
17 a Nickel-Rich Layered Li[Ni0.9Co0.05Mn0.05]O2 Cathode for Lithium-Ion Batteries. Chem.
18
19
Mater. 2017, 29 (19), 8486-8493.
20
21
22 41. Yoon, C. S.; Park, K.-J.; Kim, U.-H.; Kang, K. H.; Ryu, H.-H.; Sun, Y.-K., High-Energy
23
24 Ni-Rich Li[NixCoyMn1–x–y]O2 Cathodes via Compositional Partitioning for Next-Generation
25
26 Electric Vehicles. Chem. Mater. 2017, 29 (24), 10436-10445.
27
28
29 42. Zheng, H.; Sun, Q.; Liu, G.; Song, X.; Battaglia, V. S., Correlation between dissolution
30
31 behavior and electrochemical cycling performance for LiNi1/3Co1/3Mn1/3O2-based cells. J.
32
33 Power Sources 2012, 207, 134-140.
34
35
36
43. Kim, J.-H.; Park, K.-J.; Kim, S. J.; Yoon, C. S.; Sun, Y.-K., A method of increasing the
37
38 energy density of layered Ni-rich + ,- U1) ). )/01 cathodes (x = 0.05, 0.1, 0.2). J. Mater.
39
40 Chem. A 2019, 7 (6), 2694-2701.
41
42
44. Xiqian, Y.; Yingchun, L.; Lin, G.; Huiming, W.; Seong-Min, B.; Yongning, Z.; Khalil,
43
44
45 A.; N., E. S.; Hong, L.; Kyung-Wan, N.; Xiao-Qing, Y., Understanding the Rate Capability of
46
47 High-Energy-Density Li-Rich Layered Li1.2Ni0.15Co0.1Mn0.55O2 Cathode Materials. Adv.
48
49 Energy Mater. 2014, 4 (5), 1300950.
50
51
52 45. Croy, J. R.; Gallagher, K. G.; Balasubramanian, M.; Chen, Z. H.; Ren, Y.; Kim, D.;
53
54 Kang, S. H.; Dees, D. W.; Thackeray, M. M., Examining Hysteresis in Composite
55
56
57
58 27
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 28 of 40

1
2
3 xLi(2)MnO(3)center dot(1-x)LiMO2 Cathode Structures. J. Phys. Chem. C 2013, 117 (13), 6525-
4
5
6 6536.
7
8 46. Fu, C.; Li, G.; Luo, D.; Li, Q.; Fan, J.; Li, L., Nickel-Rich Layered Microspheres
9
10 Cathodes: Lithium/Nickel Disordering and Electrochemical Performance. ACS Appl. Mater.
11
12
13
Interfaces 2014, 6 (18), 15822-15831.
14
15 47. Bak, S.-M.; Hu, E.; Zhou, Y.; Yu, X.; Senanayake, S. D.; Cho, S.-J.; Kim, K.-B.;
16
17 Chung, K. Y.; Yang, X.-Q.; Nam, K.-W., Structural Changes and Thermal Stability of Charged
18
19
LiNixMnyCozO2 Cathode Materials Studied by Combined In Situ Time-Resolved XRD and Mass
20
21
22 Spectroscopy. ACS Appl. Mater. Interfaces 2014, 6 (24), 22594-22601.
23
24 48. Sun, Y.-K.; Lee, D.-J.; Lee, Y. J.; Chen, Z.; Myung, S.-T., Cobalt-Free Nickel Rich
25
26 Layered Oxide Cathodes for Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2013, 5 (21),
27
28
29 11434-11440.
30
31 49. Ngala, J. K.; Chernova, N. A.; Ma, M.; Mamak, M.; Zavalij, P. Y.; Whittingham, M. S.,
32
33 The synthesis, characterization and electrochemical behavior of the layered
34
35
36
LiNi0.4Mn0.4Co0.2O2 compound. J. Mater. Chem. 2004, 14 (2), 214-220.
37
38 50. Tian, C.; Nordlund, D.; Xin, H. L.; Xu, Y.; Liu, Y.; Sokaras, D.; Lin, F.; Doeff, M. M.,
39
40 Depth-Dependent Redox Behavior of LiNi0.6Mn0.2Co0.2O2. J. Electrochem. Soc. 2018, 165 (3),
41
42
A696-A704.
43
44
45 51. Lin, F.; Markus, I. M.; Doeff, M. M.; Xin, H. L., Chemical and Structural Stability of
46
47 Lithium-Ion Battery Electrode Materials under Electron Beam. Scientific Reports 2014, 4, 5694.
48
49 52. Mu, L.; Lin, R.; Xu, R.; Han, L.; Xia, S.; Sokaras, D.; Steiner, J. D.; Weng, T.-C.;
50
51
52 Nordlund, D.; Doeff, M. M.; Liu, Y.; Zhao, K.; Xin, H. L.; Lin, F., Oxygen Release Induced
53
54 Chemomechanical Breakdown of Layered Cathode Materials. Nano Lett. 2018, 18 (5), 3241-3249.
55
56
57
58 28
59
60 ACS Paragon Plus Environment
Page 29 of 40 ACS Applied Energy Materials

1
2
3 53. Liu, H.; Bugnet, M.; Tessaro, M. Z.; Harris, K. J.; Dunham, M. J. R.; Jiang, M.; Goward,
4
5
6 G. R.; Botton, G. A., Spatially resolved surface valence gradient and structural transformation of
7
8 lithium transition metal oxides in lithium-ion batteries. Phys. Chem. Chem. Phys. 2016, 18 (42),
9
10 29064-29075.
11
12
13
54. Kuppan, S.; Shukla, A. K.; Membreno, D.; Nordlund, D.; Chen, G., Revealing
14
15 Anisotropic Spinel Formation on Pristine Li- and Mn-Rich Layered Oxide Surface and Its Impact
16
17 on Cathode Performance. Adv. Energy Mater. 2017, 7 (11), 1602010.
18
19
55. Croy, J. R.; O’Hanlon, D. C.; Sharifi-Asl, S.; Murphy, M.; Mane, A.; Lee, C.-W.; Trask,
20
21
22 S. E.; Shahbazian-Yassar, R.; Balasubramanian, M., Insights on the Stabilization of Nickel-Rich
23
24 Cathode Surfaces: Evidence of Inherent Instabilities in the Presence of Conformal Coatings. Chem.
25
26 Mater. 2019, 31 (11), 3891-3899.
27
28
29 56. Tornheim, A.; Sharifi-Asl, S.; Garcia, J. C.; Bareño, J.; Iddir, H.; Shahbazian-Yassar,
30
31 R.; Zhang, Z., Effect of electrolyte composition on rock salt surface degradation in NMC cathodes
32
33 during high-voltage potentiostatic holds. Nano Energy 2019, 55, 216-225.
34
35
36
57. Lin, F.; Nordlund, D.; Markus, I. M.; Weng, T.-C.; Xin, H. L.; Doeff, M. M., Profiling
37
38 the nanoscale gradient in stoichiometric layered cathode particles for lithium-ion batteries. Energy
39
40 Environ. Sci. 2014, 7 (9), 3077-3085.
41
42
58. Lin, F.; Nordlund, D.; Pan, T.; Markus, I. M.; Weng, T.-C.; Xin, H. L.; Doeff, M. M.,
43
44
45 Influence of synthesis conditions on the surface passivation and electrochemical behavior of
46
47 layered cathode materials. J. Mater. Chem. A 2014, 2 (46), 19833-19840.
48
49 59. Yan, P. F.; Zheng, J. M.; Lv, D. P.; Wei, Y.; Zheng, J. X.; Wang, Z. G.; Kuppan, S.;
50
51
52 Yu, J. G.; Luo, L. L.; Edwards, D.; Olszta, M.; Amine, K.; Liu, J.; Xiao, J.; Pan, F.; Chen, G.
53
54 Y.; Zhang, J. G.; Wang, C. M., Atomic-Resolution Visualization of Distinctive Chemical Mixing
55
56
57
58 29
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 30 of 40

1
2
3 Behavior of Ni, Co, and Mn with Li in Layered Lithium Transition-Metal Oxide Cathode Materials.
4
5
6 Chem. Mater. 2015, 27 (15), 5393-5401.
7
8 60. Gao, A.; Sun, Y.; Zhang, Q.; Zheng, J.; Lu, X., Evolution of Ni/Li antisites under the
9
10 phase transition of a layered LiNi1/3Co1/3Mn1/3O2 cathode. J. Mater. Chem. A 2020, 8 (13),
11
12
13
6337-6348.
14
15 61. Lee, J. Y.; Kim, J. Y.; Cho, H. I.; Lee, C. H.; Kim, H. S.; Lee, S. U.; Prosa, T. J.;
16
17 Larson, D. J.; Yu, T. H.; Ahn, J.-P., Three-dimensional evaluation of compositional and structural
18
19
changes in cycled LiNi1/3Co1/3Mn1/3O2 by atom probe tomography. J. Power Sources 2018,
20
21
22 379, 160-166.
23
24 62. Duan, Y.; Yang, L.; Zhang, M.-J.; Chen, Z.; Bai, J.; Amine, K.; Pan, F.; Wang, F.,
25
26 Insights into Li/Ni ordering and surface reconstruction during synthesis of Ni-rich layered oxides.
27
28
29 J. Mater. Chem. A 2019, 7 (2), 513-519.
30
31 63. Steiner, J. D.; Mu, L.; Walsh, J.; Rahman, M. M.; Zydlewski, B.; Michel, F. M.; Xin,
32
33 H. L.; Nordlund, D.; Lin, F., Accelerated Evolution of Surface Chemistry Determined by
34
35
36
Temperature and Cycling History in Nickel-Rich Layered Cathode Materials. ACS Appl. Mater.
37
38 Interfaces 2018, 10 (28), 23842-23850.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 30
59
60 ACS Paragon Plus Environment
Page 31 of 40 ACS Applied Energy Materials

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 32 of 40

1
2
3
4
5 b)
6
a) NMC333 NMC622
7 4.0 4.0
8
Voltage (V vs Li /Li)

Voltage (V vs Li /Li)
9
+

+
10
3.6 3.6
11
12
13
3.2
14 3.2
100th
100th 1st 1st
15
16
17 0 50 100 150 200 0 50 100 150 200
18 Specific Capacity (mAh/g) Specific Capacity (mAh/g)
19
c)4.4 d)
20 NMC333 4.4 NMC622
21
22
Voltage (V vs Li /Li)

23 4.0 Voltage (V vs Li /Li) 4.0

+

24
25 3.6 3.6
26
27 100th 1st
100th 1st
28 3.2 3.2
29
30
31 0 50 100 150 200 0 50 100 150 200
32 Specific Capacity (mAh/g) Specific Capacity (mAh/g)
33 e) 200 f) 100
Discharge Capacity Retention (%)

34 180
Discharge Capacity (mAh/g)

NMC333 3-4.3V 3-4.6V

35 80
NMC622 3-4.3V 3-4.6V
36 160
37
140 60
38
39 120
40 40
41 100
42 20
80
43
44 60 0
45 0 20 40 60 80 100 0 20 40 60 80 100
46 Cycle Number Cycle Number
47
48 Figure 2. First 100-cycle charge-discharge profiles of (a, c) NMC333 and (b, d) NMC622 in the voltage window of (a, b) 3-4.3V
49 and (c, d) 3-4.6V. e) and f) show the comparison of discharge capacity and capacity retention at a cycling rate of C/10, respectively.
50
51
52
53
54
55
56
57
58 32
59
60 ACS Paragon Plus Environment
Page 33 of 40 ACS Applied Energy Materials

1
2
3
4
5
6 a) Mn K-edge
1.5
7
8

Normalized Intensity (a.u.)

9
10
11 1.0
12
13 0.10

14
15 0.5
16
0.05

17
18 0.00
6538 6540 6542 6544

19 0.0
20 6545 6550 6555 6560 6565
21
22 1.5 b) Co K-edge
23
Normalized Intensity (a.u.)

24
25
26
1.0
27
28
29 0.10
30
31 0.5
0.05
32
33
34 0.00
7708 7712
35 0.0
7705 7710 7715 7720 7725 7730 7735 7740
36
37
c) Ni K-edge
38
1.5
39
Normalized Intensity (a.u.)

40
41
42
1.0
43
44
45
46
0.5
47 NMC333 NMC622
48 Pristine
49 Cycled (4.3 V)
50 Cycled (4.6 V)
0.0
51 8335 8340 8345 8350 8355
52
Energy (eV)
53
54
Figure 3. Hard XAS K-edge spectra of pristine and cycled NMC333 and NMC622 electrodes recovered after 100 cycles in 3.0-4.3
55
V and 3.0-4.6 V: a) Mn, b) Co and c) Ni.
56
57
58 33
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 34 of 40

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
Page 35 of 40 ACS Applied Energy Materials

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 36 of 40

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
Page 37 of 40 ACS Applied Energy Materials

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 Figure 7. HAADF STEM imaging (a, b) and EELS spectra (c, d) collect on (a, c) cycled NMC333 particle and (b, d) cycled
40 NMC622 particles. The step size from the surface to the bulk is 0.5 nm. e) L2/L3 ratio of Ni, Co and Mn as well as the energy
41 difference between the pre- and main peaks of the O K-edge as a function of distance from the surface. Zone axis is [010], as shown
42 in the STEM images. Insets of (a, b) show the FFT patterns from the entire area shown in the figures.
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 37
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 38 of 40

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
Page 39 of 40 ACS Applied Energy Materials

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
 ACS Applied Energy Materials Page 40 of 40

1
2
3 Supporting Information.
4
5 List of Rietveld refinement parameters and elemental compositions from ICP measurements. EDX
6
7
mapping collected under STEM imaging. Crystallographic configurations of layered LiTMO2.
8 Rietveld refinement of synchrotron XRD patterns. HAADF STEM images and FFT analysis.
9 EELS spectra of O K-edge, Mn L-edge, Co L-edge and Ni L-edge collected on pristine and cycled
10 samples.
11
12
13
14
15 Table of Contents graphic
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 40
59
60 ACS Paragon Plus Environment