Phase Transformation Accompanied by Evolution of Internal Stress and the Coupling Mechanism of Chemical‐Mechanical Degradation in Single‐Crystal NiRich Cathodes

ABSTRACT

Operating LiNi_xCo_yMn_1‐x‐yO₂ (NCM, x ≥ 0.92) cathodes at high temperature/voltages (≥4.3 V or 45°C) to achieve high capacity inevitably leads to accelerated capacity fade. Despite extensive research into cycling behaviour under various cut‐off voltage and phase degradations, the fundamental mechanisms governing internal phase transformations, lattice deformations, and internal stress generation remain poorly understood. By using HAADF–STEM characterization with DFT and MD simulations, we disclose a new chemo‐mechanical degradation rule: lattice bending leads to the formation of O1/LiNi₂O₄ (Fd‐3m) and unstable intermediate transition phase Ni₃O₄ (Cmmm), the bending and distortion of the lattice are the direct causes of internal stress. Unlike previous findings, both RS, Ni₃O₄/LiNi₂O₄ and O1 phases were detected in various crack regions. Stress concentration from bending‐induced O1–LiNi₂O₄ and LiNi₂O₄–Ni₃O₄–RS (Fm‐3m) phase transformations leads to intracrystalline cracking, impairing capacity retention. Lattice deformation can lead to the emergence of stress and the formation of micro‐cracks, even during the O3‐O1 phase transition. This work confirmed the relationship between phase transformation and stress in the in cracked areas and stress. Meanwhile, this research provides new insights into the degradation mechanism for lithium‐ion batteries, specifically paving the way for the design and optimization of high‐energy‐density.

Voltage‐regulated chemo‐mechanical degradation of single‐crystal Ni‐rich cathodes is unraveled. Elevated voltages induce sequential O3→O1→LiNi₂O₄→Ni₃O₄→RS phase transitions via lattice bending, triggering internal stress accumulation and intracrystalline cracking. Optimized cut‐off voltage effectively stabilizes lattice integrity and Li⁺ transport.

1. Introduction

Electric vehicles (EVs), a rapidly developing mode of transportation, not only minimize the burning of fossil fuels but also have a positive impact on the environment. High energy‐density lithium‐ion batteries (LIBs) are in great demand for EVs [1, 2]. To improve the energy density of LIBs, ultra‐high Ni‐layered ternary materials (LiNi_xCo_yMn_1‐x‐yO₂, NCM, R‐3 m space group, x ≥ 0.92) have become the focus within the battery community [3]. However, the ever‐increasing Ni content in current layered cathodes considerably deteriorates their chemo‐mechanical stability, thereby posing new challenges to the commercial deployment of this material. The specific role of Ni determines the fine balance between higher capacity and safety loss due to the high reactivity of Ni and the promoted structural reconstruction on the particle surface. Further development of NCM cathodes is inevitably shifting toward larger Ni content and higher voltage, albeit at the expense of safety properties. In this context, understanding the mechanism of chemo‐mechanical degradation of layered oxide cathodes is critical for both optimizing current high‐Ni layered cathodes and designing new materials [4, 5].

One promising approach to mitigate the anisotropic volume change caused by the absence of grain boundaries is to develop single‐crystal NCM (SCNCM) with a particle size of 2–5 µm, this approach effectively suppresses the formation of microcracks. Additionally, it enhances morphology integrity and improves capacity retention compared to conventional polycrystalline secondary spheres [6]. Unfortunately, the low Li⁺ diffusion kinetics results in a relatively low reversible capacity. Recently, increasing the working temperature (≥45°C) or raising the upper cut‐off voltages (≥4.3 V) have been demonstrated as effective strategies to boost the reversible capacity of SCNCM. However, under harsh testing conditions, the high delithiation state of SCNCM inevitably leads to severe Li⁺/Ni²⁺ mixing and leaves lithium residue on the particle surface. This, in turn, readily causes an irreversible phase transition and side reactions, resulting in structural collapse of the material, particularly at high cut‐off voltages [7].

Indeed, higher Ni content combined with high voltage and elevated temperatures considerably enhances the oxidation state evolution of transition‐metal (TM) ions and triggers structural reconstruction of the crystal lattice [8]. Many studies have conducted in‐depth and detailed explorations into the relationship between structural evolution and stability. With an increase in the cut‐off voltage, the original layered phase tends to transform into the Fd‐3m spinel phase (LiNi₂O₄) or Fm‐3m rock salt (NiO, RS) phase (O3‐M‐H2‐H3‐O1‐LiNi₂O₄‐NiO), and the oxygen loss increases [9, 10, 11, 12]. Mechanical stress accumulation induced by lattice distortion is recognized as one of the core failure modes of ultra‐high nickel single‐crystal lithium‐ion batteries, and it synergistically acts with TM dissolution and electrolyte side reactions to accelerate material degradation. This stress accumulation leads to the formation of intergranular micro‐cracks, and spinel and rock salt phase transformations occur near these cracks, which are primarily caused by side reactions, especially at high voltages [13, 14]. However, the specific transformation relationships of phase transitions and the underlying mechanisms between different phase transitions, lattice distortion, and stress generation remain unclear. Additionally, the phase transition rule near microcracks requires further study. Given this background, understanding the chemo‐mechanical degradation mechanism of ultra‐high‐Ni oxides is crucial for optimizing the crystal structure of cathodes and adjusting the test voltage [15, 16].

To gain a deeper insight into the phase transition degradation mechanism in Ni‐rich SCNCM, this study examined the electrochemical behaviour and structural evolution of single‐crystal LiNi_0.92Co_0.04Mn_0.04O₂ (SC‐N92). The investigation includes the crystal structure, electronic transport, and microscopic morphology, which are closely interconnected. High‐angle annular dark field‐scanning TEM (HAADF–STEM) with super‐resolution imaging was applied to determine the redox process across different voltage ranges. This is crucial for understanding the effect of high voltage on increasing energy density and promoting structural collapse. During the electrochemical process, local lattice bending can cause stress concentration and form various structural distortion of surface and bulk kinks. We have discovered a novel phase degradation pathway for the first time, the change in stress caused by lattice bending with Li loss during cycling at different voltages can lead to transformations not only in the O3–O1 phase but also in the O1–LiNi₂O₄, LiNi₂O₄–Ni₃O₄ (Cmmm)–RS phase transformations. The appearance of O1 and LiNi₂O₄ phases in the curved zone is more likely to trigger rapid transformations of the crystal structure, indicating that lattice distortion plays a crucial role in structural evolution. Our research provides new insights into clarifying the degradation pathways of cathode materials.

2. Results and Discussion

2.1. Electrochemical Energy Storage Properties in Half‐/Full Cells of SC‐N92 Cathodes

SC‐N92 single crystals were synthesized followed by high‐temperature solid‐phase sintering adopted the precursor Ni_0.92Co_0.04Mn_0.04(OH)₂ (Canrd) (Figure S1). XRD, SEM and HRTEM (Figures S2–S5) indicated that SC‐N92 prepared by calcination of the precursors Ni_0.92Co_0.04Mn_0.04(OH)₂ has an excellent layered structure with a uniform sizes of 2–5 µm [17]. The results show that SC‐N92 particles have an α‐NaFeO₂‐type crystalline structure, which is consistent with the HAADF–STEM images shown in Figure S6, the typical interplanar spacing of 0.47 nm for the (003) plane. Additionally, EDS mapping and cross‐sectional electron probe microscopic analysis (EPMA) revealed that Ni, Co, Mn and O elements are homogeneously distributed both on the SC‐N92 surface and throughout the entire particle (Figures S7 and S8). The electrochemical performance showed that the discharge capacity of the SC‐N92 sample increased significantly with the rise of the cut‐off voltage, but there was a significant difference in cycling stability (Figure 1a). Figure 1b shows the rate capability of SC‐N92 at various voltages, showing a rapid decrease in average discharge capacities as the current density increased from 0.1 to 30 C. Notably, SC‐N92‐4.3 V exhibits an exceptionally high reversible discharge capacity of 142.59 mAh g⁻¹ within the 2.75–4.3 V range at a rate of 10 C, which represents 65.3% of the capacity measured at 0.1 C, this is higher than that observed at other voltage settings. In contrast, SC‐N92‐4.6 V exhibits a lower discharge capacity of 69.36 mAh g⁻¹ at 4.6 V, representing only 30.7% of the capacity measured at 0.1 C (A detailed explanation of SI on page 14).

FIGURE 1.

Electrochemical data in half‐/full cells for SC‐N92 at 45°C. (a) Cycling performance in half‐cells, (b) Rate performance, (c) EIS Nyquist plots before cycling and after 200 cycles within the voltage range of 4.3–4.6 V. (d) Cycling performance in pouch‐type full cells, (e) Corresponding energy density, GITT diagrams during (f) charge and (g) discharge processes, (h–j) dQ/dV contour diagrams with a cut‐off voltage window of 4.3–4.5 V at 45°C.

Figure 1c presents the EIS curves of SC‐N92 before and after 200 cycles at 45°C, the samples show an increase in R_er (R_ct before cycling) with higher voltages: 22.3, 46.1, 137.7, 259.6, and 337.9 Ω for SC‐N92, SC‐N92‐4.3, SC‐N92‐4.4, SC‐N92‐4.5, and SC‐N92‐4.6 V, respectively. This trend indicates that high‐voltage long cycles lead to enhanced polarization of SC‐N92 and weakened Li⁺ transfer kinetics. The lithium‐ion diffusion coefficient (D_Li+) value of SC‐N92‐4.3 V is significantly higher than that of the cathode at other voltages (Table S1), the difference is associated with phase transformation from the layered (O3) to the spinel/rock salt structure (RS) between the lithium layers [18]. To assess the potential practical applications, pouch‐type full batteries were assembled with graphite as the anode. These batteries were tested at 45°C with cut‐off potentials of 4.3, 4.4, and 4.5 V (Figure 1d). Notably, the SC‐N92‐4.3 V cell maintained a reversible capacity of 189.7 mAh g⁻¹ after 300 cycles, demonstrating an exceptional capacity retention of 89.2%. In contrast, the SC‐N92‐4.4 and SC‐N92‐4.5 V exhibited considerably lower performance, with discharge capacities dropping to 122.4 and 66.5 mAh g⁻¹ after 240 cycles, respectively. The corresponding capacity retentions were 56.8% and 30.3% (Figure S9a–c). Importantly, SC‐N92‐4.3 V demonstrated an exceptionally high coulombic efficiency of approximately 99.95%. In contrast, the SC‐N92‐4.5 V cathode showed a gradually deteriorating coulombic efficiency (Figure S9d). Moreover, based on the middle voltage (Figure S9e), the specific energy density of SC‐N92‐4.3 V reached 700.9 Wh kg⁻¹, which is considerably higher than that of SC‐N92‐4.5 V (211.5 Wh kg⁻¹ after 240 cycles, Figure 1e). Meanwhile, the GITT test was used to further evaluate the influence of phase transition on the lithium‐ion transport kinetics, the value of D_Li+ showed a strong dependence on the cut‐off voltage. During the initial stage of charging and discharging, the D_Li+ values of all samples fluctuated between 10⁻¹⁰ and 10⁻¹². However, SC‐N92‐4.3 V maintained a relatively high diffusion coefficient, with the average value remaining at around 10⁻¹⁰ (Figure 1f). This indicates that the two‐dimensional lithium‐ion transport channels remain unobstructed within the stable layered O3 phase, which is conducive to the rapid migration of Li⁺. In contrast, as the cut‐off voltage rose to 4.4 and 4.5 V, the D_Li+ value dropped sharply (Figure 1g). This severe dynamic difference is attributed to the variation in the degree of phase transformation. The differential capacitance (dQ/dV) were calculated from the discharge curves, the peak position and intensity of SC‐N92‐4.3 V remains nearly constant throughout, indicating minimal electrode polarisation and capacity decay [19]. Additionally, SC‐N92‐4.3 V exhibits a higher voltage platform for the H2–H3 phase transition (Figure 1h; Figure S10a). In contrast, the peak of SC‐N92‐4.4 V gradually shifts to a lower voltage range, accompanied by a decrease in intensity, indicating severe polarisation (Figure 1i; Figure S10b). Moreover, the phase transition of SC‐N92‐4.5 V clearly decreases in intensity over the cycles, this is the primary cause of structural collapse and subsequent crack formation (Figure 1g; Figure S10c).

2.2. Structural Transformations and Underlying Reaction Mechanism

The intuitive observation of microstructure evolution and the quantitative analysis of macroscopic mechanical properties aim to reveal the mechanical failure behavior of battery under specific working conditions (different voltages). SEM and cross‐section SEM images (Figure S11) clearly show that the depth of cracks within the entire particle increases with higher voltages, a phenomenon attributed to the concentration of internal stress caused by deep lithiation and delithiation processes. Continuous FIB‐SEM analysis (Figure S12) further demonstrated that intracrystalline cracks occurred at random locations within the crystal (Figure 2a,b). High‐resolution TEM images further show that the SC‐N92‐4.5 V electrode exhibits severe surface degradation, accompanied by a clear structural transition from the layered O3 to the rock‐salt phase, the phase transition process severely hinders the transport of Li⁺ (Figure S13) [20]. At lower voltages (4.4 V), this detrimental phenomenon is suppressed (Figure S14). Furthermore, the surface degradation is substantially improved, and no noticeable phase transition is observed in the SC‐N92‐4.3 V electrode (Figure S15). These findings confirm that the damage degree of surface to the material increases with higher cut‐off voltages. Additionally, HAADF–STEM imaging obtained along the [100] zone axis, combined with the focused ion beam (FIB) technique, was employed to analyse the structural evolution. As illustrated in Figure 2c, only a minor TM migration and a partially disordered structure (Figure 2cI‐1) were observed on the SC‐N92‐4.3 V surface, this observation is corroborated by a slight reduction in the intensity of the TM layer (Figure S16I‐1). Meanwhile, the slight lattice distortions occur in a mixed phase transition from O1 to LiNi₂O₄ (Fd‐3m) spinel (Figure 2cI‐2; Figure S16I‐2), similar lattice bending phenomena were detected in various regions of the interface, providing evidence that the phase transition process is accompanied by lattice distortion (Figure S17).

FIGURE 2.

Morphology and crystal structure for cathode in pouch‐type full cells at 45°C after 300 cycles. Continuous FIB–SEM of single particles with different sections for (a) SC‐N92‐4.3 V and (b) SC‐N92‐4.5 V. HAADF–STEM of (c) surface and (d) 200‐nm‐deep positions for SC‐N92‐4.3 V, (e) surface and (f) 200‐nm‐deep positions for SC‐N92‐4.4 V, (g) surface and (h) 200‐nm‐deep positions for SC‐N92‐4.5 V with corresponding strain mapping using GPA patterns and enlarged lattice images.

Moreover, twisted mixed phases with varying degrees of phase transition were observed from the upper surface to a depth of 200 nm (Figure 2d; Figure S18 in ABF). The lattice bending‐induced chemo‐mechanical phase transition was found to occur not only in the O3–O1 transition (Figure 2dII‐1; Figure S19II‐1) but also at O1‐LiNi₂O₄ (Figure 2dII‐2; Figure S19II‐2) [21]. Importantly, this bending phenomenon differs from the O3–O1 phase transition bending previously reported in the literature, as it represents the intermediate stage transformation process of the crystal structure undergoing the O3‐LiNi₂O₄ phase transition. Furthermore, the GPA simulation patterns based on HAADF–STEM show the corresponding lattice stress distribution [22]. GPA analysis shows that stress concentrations vary in different phase transformation regions, with notably higher stress levels observed within the bending areas (Figure 2cI). Notably, stress concentrations are present in the lattice deformation zones associated with both O3–O1 and O1–LiNi₂O₄ phase transitions at the 200‐nm depth, confirming a direct correlation between stress generation and lattice distortion (Figure 2dII) [23].

Upon examination, a mixed phase consisting of ordered RS phase, disordered RS, Ni₃O₄ (Cmmm) and LiNi₂O₄ (Fd‐3m) was identified on the SC‐N92‐4.4 V surface following 300 cycles (Figure 2e). The degree of Li/TM intermixing determines the ordered and disordered degree of the RS lattice, lattice bending and partial Li loss, leading to partial Li/TM intermixing and the formation of disordered RS with lattice defects—this process is driven by moderate stress concentration induced by lattice distortion. When SC‐N92 experiences complete Li loss and continuous lattice reconstruction, TM ions form a long‐range ordered distribution in the lattice, finally forming the perfect ordered RS phase. Meanwhile, Ni₃O₄ (Cmmm) is confirmed as a key intermediate phase in the LiNi₂O₄ spinel‐to‐RS phase transition, Ni₃O₄ (Cmmm) is an orthorhombic, spatial group Cmmm nickel oxide structure, commonly found as a defect/disordered phase transformation product after high‐voltage lithiation, and is not a stable spinel or rock salt phase. Its characteristics are layered distortion, severe cation mixed arrangement and lack of unobstructed Li⁺ channels.

Lattice bending phenomena were also observed between the composite phases, further indicating a direct correlation between the phase transition process and lattice bending. Compared with the O3 layered structure with clear Li⁺ transport channels, the Ni₃O₄ and RS phase has no regular Li⁺ insertion sites and ordered transport channels. Li⁺ can only migrate through paths with extremely high diffusion energy barriers or phase defect locations, resulting in a significant inhibition of Li⁺ transport (Figure 2eI‐1, I‐2; FigureS20I‐1, I‐2). HAADF–STEM revealed the presence of intracrystalline cracks at 200 nm from the top surface. Contrary to previous research findings, no substantial RS phase was detected on either side of the crack. Instead, LiNi₂O₄ and Ni₃O₄ phases were observed accompanying the crack, indicating that prolonged lattice distortion during phase transition inevitably leads to stress generation and concentration (Figure 2fII‐1, II‐2; Figure S21II‐1, II‐2). When localized regions within the particle are unable to withstand the accumulated stress, intracrystalline cracks form and progressively expand as stress levels continue to rise. GPA simulations showed a considerable stress distribution within the phase transition bending regions. When comparing the surface region to the LiNi₂O₄–RS phase transition bending area (Figure 2eI), a notably uniform stress distribution was observed within the internal crack region, characterized by minimal lattice distortion. This finding further substantiates the correlation between stress levels and lattice bending (Figure 2fII) [24]. Due to the limitation of HAADF‐STEM viewing area, further characterization of the crack region reveals extensive lattice bending, which confirms that lattice bending induces phase transition and internal stress accumulation (Figure S22).

Conversely, SC‐N92‐4.5 V samples exhibited a substantial presence of both severely ordered and disordered RS phases distributed across the particle surfaces, despite exhibiting only slight lattice distortion. The transition from the particle surface to the interior within the RS structure was accompanied by a notably high intensity of the Li/TM mixing peak (Figure 2gI‐1, I‐2; Figure S23I‐1, I‐2). Compared with the O3 layered structure with long‐range ordered Li⁺ intercalation sites and 2D transport channels, the disordered RS phase has a face‐centered cubic structure with no regular Li⁺ accommodation sites, resulting in the loss of effective Li⁺ transport channels [25]. At high voltages, an examination of deeper cracks within the crystal revealed a predominance of both severely ordered and disordered RS phases lining both sides of the cracks. Importantly, it has been established that during the O3–RS phase transition, there is a direct correlation between the extent of phase transformation and the accumulation of internal stress. Furthermore, the presence of severe internal stress can trigger both the initiation and propagation of cracks, affecting the diffusion of Li⁺ and further damaging the crystal structure (Figure 2hII‐1, II‐2; Figure S24II‐1, II‐2). GPA simulations showed a substantial amount of stress generation and concentration within the bending region, particularly at the phase transition site where disordered RS transforms into ordered RS (Figure 2gI). In contrast, no significant stress generation or accumulation was observed in the uncurved RS phase, which exhibited a uniform stress distribution at 200 nm within a region characterized by the absence of lattice distortion (Figure 2hII) [26].

Rietveld refinement results further prove the existence of each phase, pristine SC‐N92 shows a pure O3 layered structure without impurity phases (Figure S25a). SC‐N92‐4.3 V still maintains the dominant O3 phase with only minor lattice distortion, while secondary phases (Fd‐3m LiNi₂O₄, Cmmm Ni₃O₄ and Fm‐3m rock salt) occupy a low proportion (Figure S25b). By contrast, SC‐N92‐4.4 V triggers obvious multiphase evolution (Figure S25c). The O3 phase peaks markedly weaken, accompanied by evident signals of LiNi₂O₄, Ni₃O₄ and rock‐salt phases, revealing irreversible structural deterioration. The layered structure suffers severe degradation for SC‐N92‐4.5 V, O3 peaks further attenuate and broaden, whereas secondary phases especially the inert rock‐salt phase are greatly intensified (Figure S25d and Table S2), suggesting continuous evolution toward irreversible electrochemically inert phases with elevated voltage. To further investigate the phase transitions occurring in the electrode after cycling, we used XRD pole figures to analyse the exposure of the corresponding crystal planes. Severe irreversible phase transitions result in a reduced number of Li accommodation sites and an increased degree of structural disorder. Within the lower voltage of 4.3 V, lattice distortion is minimal and accompanied by a uniform distribution of Li‐ions. The consistent orientation of the (003), (101), and (006)/(012) planes maintains the integrity of the crystal structure (Figure 3a). Conversely, the substantial lattice changes at high voltage (4.5 V) impair structural integrity, the alignment of the (003), (101), and (006)/(012) planes exhibits poor consistency (Figure 3b). These findings directly demonstrate that RS and electrochemically inactive Li⁺ are more prone to forming on particle under high‐voltage conditions [27].

FIGURE 3.

XRD poles molecular dynamics simulation for cathode in pouch‐type full cells at 45°C after 300 cycles. XRD poles depicting the (003), (101), and (006)/(012) lattice planes for (a) SC‐N92‐4.3 V and (b) SC‐N92‐4.5 V. Molecular dynamics simulation plots for the systems of (c) SC‐N92‐4.3 V, (d) SC‐N92‐4.4 V and (e) SC‐N92‐4.5 V.

Furthermore, to elucidate the impact of lattice distortion on the overall system, MD simulations were employed to model specific Li⁺ transport mechanisms for various phase transitions. Importantly, the model is constructed based on HAADF‐STEM to ensure the accurate representation of the experiment. Following the formation of various complex phases, which are accompanied by lattice distortion and alterations in crystal structure, the lattice mismatch between different phases modifies the Li⁺ transport mode and channels. Notably, based on the O3–O1 phase transition and the subsequent O1‐LiNi₂O₄, the calculated diffusion coefficients for all phases are presented in Table S3 and Movie S1 [28], revealing that the O3 structure exhibits the highest performance (rate = 0.55 Å/ps). Additionally, the Li⁺ migration barrier in the O3–O1 bending region is low, resulting in minimal impact on the Li⁺ transport rate. The crystal deformation process is accompanied by the formation of spinel LiNi₂O₄ (Fd−3m space group), which alters both the Li⁺ content within the lattice and the ordered Li⁺ transport channels of the original layered structure. The coexistence of different phase structures modifies the Li⁺ transport channels, reducing the transport rate (rate = 0.146 Å/ps) as Li⁺ migrates into the LiNi₂O₄ structure over time (Figure 3c). Lattice expansion and deformation intensify lead to structural bending in the SC‐N92‐4.4 V material [29]. Furthermore, as the Li⁺ transport channels within the crystal decreases, the diffusion rate of Li⁺ ions becomes substantially restricted. Theoretically, the Li‐deficient RS phase represents the final degradation product resulting from lattice distortion in the spinel phase. Unlike the typical rapid Li⁺ diffusion through 2D channels, both Ni₃O₄ and RS phases lack distinct Li⁺ transport pathways. Consequently, Li⁺ can only migrate via phase defects, and the elevated diffusion energy barrier impedes the transport rate of Li⁺ (Movie S2). When Li⁺ traverse the Ni₃O₄ spinel and disordered RS phases, the transport is considerably impeded, clearly demonstrating the direct impact of lattice bending following the phase transition (Ni₃O₄–RS) (Figure 3d). Notably, a large number of mixed phases related to spinel and RS are distributed on the crystal for SC‐N92‐4.5 V, which are accompanied by various structures configurations and defects, ultimately form an ordered RS phase. As internal Li⁺ migrates toward the chaotic structure of the Ni₃O₄ and ordered RS, Li⁺ are basically not transported in Ni₃O₄ (rate = 0.02818 Å/ps), while ordered RS phases completely lack Li⁺ transport channels, resulting in failed Li⁺ diffusion (Movie S3; Figure 3e). This phenomenon leads to a rapid degradation of the discharge specific capacity of the cathode electrode.

2.3. In Situ Mechanism for Redox Mechanism, Concentration and Stress Simulations

In‐situ XRD measurements were performed to investigate the structural evolution and reaction reversibility at various voltages. The phase transition correlates with a voltage distribution, where the (003) peak gradually shifts toward lower 2θ angles as the charge reaches 4.1 V. This movement indicates a gradual expansion of the c‐axis due to changes in electrostatic repulsion, transitioning from the H1 to M phase. Subsequently, during the H2–H3 phase transition, the c‐axis parameter undergoes a drastic contraction, causing the (003) peak to continuously shift to higher 2θ angles at elevated voltages [30]. The (003) diffraction peak of SC‐N92‐4.3 V shifts to the right by 0.84° during the H2–H3 transition (Figure 4a,d), which is notably smaller than the 1.32° rightward shift observed for SC‐N92‐4.5 V (Figure 4b,e). Upon full charging to 4.6 V, the (003) diffraction peak shifts sharply in the 2θ direction, indicating a severe lattice contraction along the c‐axis of 1.68° during the H2–H3 phase transition (Figure 4c,f). Furthermore, as shown in Figure 4g, the contraction rate of the c‐parameter for SC‐N92‐4.3 V (3.91%) is considerably lower compared to SC‐N92‐4.5 V (4.82%) and SC‐N92‐4.6 V (6.41%). During the discharge process, the (003) diffraction peak returns to original 2θ angle position, indicating that the reaction is a highly reversible lithiation/delithiation process [31]. These findings demonstrate that maintaining a lower voltage of 4.3 V effectively preserves lattice integrity, thus mitigating structural degradation and minimizing electrode polarisation with minimal changes in (003) reflection. While increasing the cut‐off voltage can enhance specific discharge capacity, it inevitably results in structural degradation and a severe H2–H3 phase transition. Consequently, optimizing the cut‐off voltages is crucial for enhancing overall battery performance.

FIGURE 4.

In situ XRD analysis in the first cycle and COMSOL in pouch‐type full cells at 45°C after 300 cycles. In‐situ XRD of SC‐N92 cathodes at (a) 2.75–4.3 V, (b) 2.75–4.5 V and (c) 2.75–4.6 V. Local amplification of the (003) diffraction peak at (d) 2.75–4.3 V, (e) 2.75–4.5 V and (f) 2.75–4.6 V. (g) The variation in the c‐axis lattice parameter during deintercalation. Li⁺ concentration distributions for (h, i) SC‐N92‐4.3 V and (j,k) SC‐N92‐4.5 V of adjacent particles; Stress simulation diagrams for (l,m) SC‐N92‐4.3 V and (n, o) SC‐N92‐4.5 V of adjacent particles. Li⁺ concentration distributions for (p) SC‐N92‐4.3 V and (q) SC‐N92‐4.5 V of dispersed particles; Stress simulation diagrams for (r) SC‐N92‐4.3 V and (s) SC‐N92‐4.5 V of dispersed particles.

The distribution of Li⁺ concentration and stress (both internal and particle collision‐related) was simulated and analysed using COMSOL software after 300 cycles. During the lithiation process for SC‐N92‐4.3 V, the concentration of Li⁺ inside the particles increases, resulting in a relatively uniform distribution with a small gradient due to the mild phase transition. In this scenario, Li⁺ are primarily transported through two‐dimensional radial channels [32]. During the delithiation state, Li⁺ undergo uniform deintercalation from within the cathode particle, resulting in the final particles exhibiting the lowest lithium‐ion concentration distribution (Figure 4h; Movie S4). At higher cut‐off voltages (4.5 V), the disparity in phase transformation leads to the formation of a substantial amount of irregular Ni₃O₄ and RS phases on the cathode particle surface, which alters the lithium ion transport channel. The severe phase transition induces a substantial gradient in lithium ion concentration and an uneven concentration distribution throughout the delithiation/lithiation process following 300 cycles (Movie S5), resulting in the internal particle having a higher lithium ion concentration than the surface (Figure 4j). To illustrate the concentration distribution differences more clearly, a 3D concentration diagram was simulated. Compared with SC‐N92‐4.3 V (Figure 4i), the SC‐N92‐4.5 V shows a notably larger Li⁺ concentration difference and gradient from the surface to the interior bulk (Figure 4k) [33]. The long‐term cycle will lead to irreversible phase changes, which will impede the transport of Li⁺ and result in subtle concentration differences at the beginning of discharging and the end of charging.

Severe phase transitions not only impede the Li⁺ diffusion but also trigger the generation and accumulation of mechanical stress within particles. The strain and stress show remarkable similarity during delithiation/lithiation process, indicating excellent strain reversibility for SC‐N92‐4.3 V, the Li⁺ concentration distribution remains relatively uniform with minimal stress accumulation (Movie S6). Meanwhile, single crystal particles undergo expansion and contraction, which may involve extrusion and collision between neighbouring particles. Importantly, the stress concentration in SC‐N92‐4.3 V barely causes any damage to the original structure (Figure 4l). In contrast, the SC‐N92‐4.5 V electrode exhibits poor Li⁺ diffusion characteristics with the continuous increase of equivalent stress. These large and non‐uniform stress distributions are sufficient to induce irreversible structural damage to the SC‐N92 electrode (Movie S7). Simultaneously, the stress phenomenon resulting from particle extrusion is clearly observable (Figure 4n). The corresponding 3D stress simulation reveals a similar pattern: compared to the low stress for SC‐N92‐4.3 V electrode (Figure 4m), both interface and internal stress concentrations are considerably more pronounced in the SC‐N92‐4.5 V electrode (Figure 4o).

After 300 cycles, the simulation was extended to dispersed multi‐particle systems, which exhibit evolutionary trends consistent with those observed in adjacent particles. The multi‐particle of SC‐N92‐4.3 V in a dispersed state maintains a relatively uniform lithium‐ion concentration distribution during the lithiation and delithiation processes, characterized by a minimal concentration gradient (Figure 4p; Figure S26a). In contrast, the dispersed particles of SC‐N92‐4.5 V display pronounced concentration disparities (Figure 4q; Figure S26b). The formation of spinel/rock‐salt phases on the surface severely impedes internal lithium‐ion transport, leading to significant concentration heterogeneity throughout the dispersed particle clusters. Furthermore, stress simulation results substantiate the mechanical failure mechanisms within these dispersed systems. SC‐N92‐4.3 V demonstrates excellent strain reversibility with negligible inter‐particle stress (Figure 4r; Figure S27a), the dispersed SC‐N92‐4.5 V generates more substantial equivalent stress accumulation (Figure 4s; Figure S27b). Unlike the stress concentrations primarily driven by particle extrusion in adjacent systems, the non‐uniform stress distribution in these dispersed assemblies is predominantly governed by intracrystalline lattice distortion, which remains sufficient to trigger structural degradation and crack propagation.

To investigate electron/ion transport ability and charge compensation mechanisms at different voltages, DFT calculations and x‐ray absorption near‐edge structure (XANES) spectroscopy were used to assess SC‐N92 electrode after 300 cycles. Based on HAADF–STEM results, different crystal structures were modelled at the same interface position under varying voltages. Energy profiles and barrier diagrams indicated that the lowest migration barrier was along the O3–O1 path with a lower phase transition degree, where Li⁺ can easily diffuse through the 2D migration path (Figure 5a). During O1–LiNi₂O₄ and Ni₃O₄–RS phase transitions, the decrease in Li⁺ transport channels and increase in energy barriers greatly considerably hinders the Li⁺ transport (Figure 5b–d). Analysis of 2D/3D differential charge density revealed a pronounced electron density accumulation around oxygen atoms, while TMs exhibit a characteristic electron‐deficient state. Notably, interlayer electronic interactions are weaker compared to other systems in SC‐N92‐4.3 V (with a smaller interlayer iso‐surface range), and the increased layer spacing facilitates easier Li⁺ migration (Figure 5e).

FIGURE 5.

DFT calculations and chemical characterisations in pouch‐type full cells at 45°C after 300 cycles for different voltages. The crystalline structures for (a) SC‐N92‐4.3 V, (b) SC‐N92‐4.4 V, and (c) SC‐N92‐4.5 V cathodes and (d) energy profiles of Li⁺ migration pathways; 3D and 2D differential charge density distributions for (e) SC‐N92‐4.3 V, (f) SC‐N92‐4.4 V and (g) SC‐N92‐4.5 V cathodes; The total and partial density of states plots for (h) SC‐N92‐4.3 V, (i) SC‐N92‐4.4 V, and (j) SC‐N92‐4.5 V cathodes; (k) XANES spectra of Ni K‐edge in the fully charged state; (l,m) The corresponding local magnifications; (n) Ni K‐edge EXAFS at different voltages; 2D Fourier transformed EXAFS spectra for (o) SC‐N92‐4.3 V, (p) SC‐N92‐4.4 V, and (q) SC‐N92‐4.5 V.

The local interatomic bond cooperation is enhanced for SC‐N92‐4.4 V, the O1 and LiNi₂O₄ phase interface with smaller spacing is formed, which consequently increases interlayer diffusion resistance (Figure 5f). Additionally, the Ni₃O₄–RS transition phase interface developed a robust TM─O bond for SC‐N92‐4.5 V, evident from the pronounced red and blue colour changes, which extend across the entire interface region. As layer spacing further diminishes and interfacial binding strengthens, the strong interfacial binding facilitates crystal stability, resulting in a corresponding reduction of Li storage ability (Figure 5g). Compared to the SC‐N92‐4.3 V system (Figure 5h), the density of states at the Fermi level in the SC‐N92‐4.4 V system significantly decreased (Figure 5i), indicating diminished electron conductivity. The pseudoenergy gap in the spin‐down portion on both sides of the Fermi level becomes more pronounced due to the enhanced of phase boundary bonds, which also inhibits interlayer Li⁺ diffusion [34]. The extent of DOS peak broadening continues to increase, indicating the formation of stronger interatomic electron interactions, which enhance the overall bonding strength of the system [35]. The strengthening of the bond further reduces Li⁺ migration ability for SC‐N92‐4.5 V (Figure 5j).

The hard x‐ray absorption near‐side structure (XANES) technology was used to analyse the chemical valence states of Ni (charge compensation) under fully charged conditions after 300 cycles. Theoretically, higher charged cut‐off voltage lead to more intense electrochemical reactions of Ni²⁺ to Ni³⁺/Ni⁴⁺, causing the Ni to shift to a higher energy range (Figure 5k) [36]. Notably, Ni K‐edge positions in the XANES spectra differ among the cathodes: SC‐N92‐4.3 V (8344.7 eV) has a higher position compared to SC‐N92‐4.4 V (8343.8 eV) and SC‐N92‐4.5 V (8343.1 eV). This indicates that SC‐N92‐4.3 V has the highest overall average oxidation state, corresponding to a higher content of Ni³⁺/Ni⁴⁺ during continuous reversible redox reactions. Notably, SC‐N92‐4.5 V displays lower energy with a reduced Ni³⁺/Ni⁴⁺ content compared to SC‐N92‐4.3 V, which is likely attributed to less content of active Ni²⁺ on the interface caused by the irreversible phase transformation. (Figure 5l,m) [37]. As shown in Figure 5n, the Fourier transform (FT) magnitudes of the extended X‐ray absorption fine structure (EXAFS) spectrum at the Ni K‐edge exhibit two prominent peaks. Notably, compared to the SC‐N92‐4.5 V, the SC‐N92‐4.3 V demonstrates a significantly enhanced strength of the main Ni–O shell, attributed to the formation of a greater amount of Ni⁴⁺ upon Li⁺ release. Interestingly, the disappearance of peaks associated with the first subshell indicates that the majority of Ni²⁺ ions were oxidized to Ni³⁺ and subsequently to Ni⁴⁺. However, the SC‐N92‐4.5 V sample exhibited a shorter length of the corresponding Ni−O bond, suggesting reduced oxidation of Ni²⁺ and confirming that the SC‐N92‐4.5 V undergo substantial phase transitions and crystal collapse under high‐voltage conditions. To analyse the local environment of Ni, EXAFS wavelet transform analysis was performed [38]. As illustrated in Figure 5o, the peak shape and average distance of the Ni–O and Ni–M peaks for SC‐N92‐4.3 V closely resemble those reported in the literature, indicating a typical O3 structure with only minor phase transitions and degradation [39]. In contrast, the crystal structures of SC‐N92‐4.4 V and SC‐N92‐4.5 V underwent major alterations, as evidenced by their irregular Ni–O and Ni–M peak shapes and bond lengths (Figure 5p,q). Consequently, the chemical environment of these materials deviates from the typical layered structure.

Time‐of‐flight secondary‐ion mass spectrometry (TOF–SIMS) analysis was used to evaluate the structural stability of SC‐N92‐4.3 V and SC‐N92‐4.5 V after 300 cycles. The detrimental electrolyte/electrode interface reaction exacerbates particle degradation, involving the formation of LiNi₂O₄/Ni₃O₄–RS, TM layer dissolution, and lattice oxygen loss. Specifically, side reactions cause TM degradation and the formation of transition metal fluorides (e.g., NiF₃ ⁻, CoF₃ ⁻ and MnF₂ ⁻), the concentration distribution and 3D render demonstrate that the distributions of NiF₃ ⁻, CoF₃ ⁻ and MnF₂ ⁻ from the surface to the interior of the SC‐N92‐4.3 V cathodes are notably lower and more uniform compared to those of the SC‐N92‐4.5 V cathode (Figure 6a,b; Figure S28). The corresponding depth profile curve (Figure 6c) confirms that acid corrosion by‐products (such as HF) in the electrolyte exhibit greater activity at higher voltages. P compounds (e.g., PF₂O₂ ⁻) and organic species (e.g., CF⁻, CN⁻) primarily originate from the decomposition of solvents and salts in the electrolyte, these components tend to distribute from the outer surface toward the interior of the CEI film. The degradation‐induced fragments on SC‐N92‐4.3 V are much lower compared those on SC‐N92‐4.5 V, as evidenced by the cumulative signals of PF₂O₂ ⁻, CF⁻ and CN⁻ from the concentration distribution and 3D render analysis (Figure 6d–e) [40]. Notably, SC‐N92‐4.5 V exhibits instability, where organic electrolytes induce degradation of the surface layer structure, the decomposition of solvents and salts in the electrolyte is considerably suppressed by controlling the cut‐off voltage (Figure 6f) [41].

FIGURE 6.

Characterisation of the morphology in pouch‐type full cells at 45°C after 300 cycles. The TOF‐SIMS concentration distribution and 3D rendering of the composition (NiF₃ ⁻, CoF₃ ⁻, and MnF₂ ⁻) of TM phase for (a) SC‐N92‐4.3 V and (b) SC‐N92‐4.5 V with (c) corresponding depth profile curve; the concentration distribution and 3D rendering of the composition (PF₂O₂ ⁻, CF⁻ and CN⁻) of CEI film for (d) SC‐N92‐4.3 V and (e) SC‐N92‐4.5 V with (f) corresponding depth profile curve. Depth XPS contour diagram of C 1s, O 1s, F 1s for (g) SC‐N92‐4.3 V and (h) SC‐N92‐4.5 V within the 0–300 nm range. A single‐particle force test for (i) SC‐N92‐4.3 V and (j) SC‐N92‐4.5 V; (k) Curve graph for the single‐particle force test. TEM images of the cathode for (l) SC‐N92‐4.3 V and (m) SC‐N92‐4.5 V; TEM images of the anode for (n) SC‐N92‐4.3 V and (o) SC‐N92‐4.5 V; LA–ICP–MS with Ni, Li, and P distributions for (p) SC‐N92‐4.3 V and (q) SC‐N92‐4.5 V.

Additionally, the chemical composition and environment were analysed using XPS from the surface (0 nm) to the interior (300 nm), revealing the decomposition of carbonates (ROCO₂Li, ROLi and Li₂CO₃) and binders/conductive additives in the electrolyte solvents. The CO₃ ²⁻ concentration in C 1s, O 1s peaks for the SC‐N92‐4.3 V electrode gradually decreases from the surface (0 nm) to the interior (300 nm) of the particles, indicating that the particles maintain their integrity well, with limited electrolyte penetration into the particle structure (Figure 6g). Additionally, with increasing depth, the peak intensities of Li–F (approximately 684 eV) and C–F (approximately 687 eV) in the F 1s spectrum decrease substantially (Figure S29), demonstrating that the SC‐N92‐4.3 V electrode effectively inhibits the decomposition of the LiPF₆ electrolyte, which stabilizes the layered structure and suppresses the surface electrode/electrolyte reaction of SCNCM [42]. Notably, the peak intensities of CO₃ ²⁻ in the C 1s and O 1s spectrum of the SC‐N92‐4.5 V electrode are significantly stronger compared to those of SC‐N92‐4.3 V electrode, confirming increased breakdown of carbonate solvents at the SCNCM/electrolyte interface under high voltage (Figure 6h) [43]. Meanwhile, the strength peaks of Li–F and C–F of SC‐N92‐4.5 V electrode are relatively strong, indicating that the interface was severely corroded (Figure S30). Specifically, both depth‐profiled XPS (F 1s) and TOF‐SIMS (PF₂O₂ ⁻, CF⁻) analyses reveal that the CEI formed for SC‐N92‐4.5 V displays markedly higher fluorine intensity, deeper fluorine penetration, and a steeper fluorine concentration gradient relative to the SC‐N92‐4.3 V. This distinct difference directly confirms that elevated voltage accelerates the hydrolysis and decomposition of the LiPF₆ ⁻ based electrolyte, generating abundant fluorine‐containing byproducts and promoting the formation of a thick and unstable CEI layer. In contrast, the moderate voltage (4.3 V) effectively mitigates electrolyte degradation, enabling the formation of a thin, uniform and stable CEI film.

To investigate the crystal strength of the particles, in‐situ optical pressure tests were conducted on SC‐N92‐4.3, SC‐N92‐4.4, and SC‐N92‐4.5 V particles, the SC‐N92‐4.3 V cathode exhibited a longer breaking time and a lower degree of particle damage throughout the entire process (Figure 6i, Movie S8), indicating that SC‐N92‐4.3 V has a superior compressive resistance under low‐voltage conditions and is able to maintain structural integrity more effectively during prolonged cycling (Figure S31). High voltage (4.4 or 4.5 V) can induce severe lattice expansion and internal stress accumulation in the cathode material (SC‐N92), making it more prone to cracking and rapid propagation under mechanical pressure (Figure 6j, Movies S9–S10), thereby significantly reducing mechanical strength (fracture toughness, Figures S32 and S33). The force‐displacement curve can indirectly reflect the toughness ability to resist fracture. The curve of the SC‐N92‐4.3 V is significantly “higher”, indicating that its hardness and modulus have been comprehensively enhanced (Figure 6k). TEM revealed that a dense and uniform CEI film formed on the cathode surface of SC‐N92‐4.3 V (Figure 6l). In contrast, the surface roughness of SC‐N92‐4.5 V particles slightly increased with a thick and uneven CEI film, leading to noticeable changes in the surface morphology (Figure 6m). While the cathode electrode undergoes degradation, a series of changes also occur on the anode electrode side of the graphite. TEM shows that there is an excellent thin and uniform SEI film on the surface of SC‐N92‐4.3 V (Figure 6n), while the surface of SC‐N92‐4.5 V is an inferior thick and uneven SEI film (Figure 6o). The thin and uniform SEI film corresponds to very few harmful metal impurities (Ni), side reaction products (P) and the uniform distribution of Li⁺, indicating an efficient, stable electrode interface with few side reactions of SC‐N92‐4.3 V. The unstable and continuously growing SEI film of SC‐N92‐4.5 V leads to an increase in ion migration resistance and internal resistance, constantly consuming active lithium and electrolyte, resulting in rapid capacity decline and shortened lifespan.

Trace elements at the anode electrode were analysed using laser ablation inductively coupled plasma mass spectrometry (LA–ICP–MS). The TM (Ni) dissolved at the cathode electrode and was transported through the electrolyte to the anode electrode, where it was deposited. Notably, the Ni element intensity distribution in SC‐N92‐4.3 V was both uniform and characterized by a low overall content, confirming the presence of a mild electrochemical reaction at low voltage and the uniform composition of the SEI layer (Figure 6p). In contrast, the phase transition in SC‐N92‐4.5 V is highly intense, leading to the dissolution of the TM. The strong signal of Ni indicates significant damage to the cathode material and the formation of an uneven SEI film on the anode (Figure 6q). Additionally, the observed differences in the elemental distribution of Li and P further substantiate the mechanism of SEI formation on the anode electrode as a function of voltage. Importantly, the mild electrochemical reaction helps prevent simultaneous damage to both the cathode and anode electrodes by the electrolyte.

The investigation of the underlying reaction mechanism has elucidated the structural evolution of SC‐N92 under various high cut‐off voltages. Generally, the delithiation/lithiation process involves a phase transition from O3 to O1, followed by the formation of O1‐LiNi₂O₄/ Ni₃O₄–RS transition, with these transformations becoming more pronounced as the cut‐off voltage increases.

At lower voltages (4.3 V), the phase transition is relatively mild, accompanied by minimal lattice deformation, which can inhibit the continuous accumulation of mechanical stress and the appearance of cracks. Consequently, only a minor degree of disorder manifests on the particle surface. The detailed lattice diagram reveals that the O1 phase predominantly resides in the curved regions, which are more susceptible to oxygen loss. Compared to the O3 phase, the O1 phase is more energetically conducive to the development of the spinel or RS phase for the TM layer to the Li layer migration (Figure 7a). Additionally, the complex phases exist near and on the surface and the cracks in SC‐N92‐4.4 V, including LiNi₂O₄ and Ni₃O₄, as well as disordered RS phases. Ni₃O₄ serves as an intermediate phase for the transition to disordered RS phases (Figure 7b). Overall, the presence of complex phases hinders the diffusion of lithium ions, resulting in a Li⁺ concentration gradient and causing severe tensile strain. In contrast, intercrystalline cracks in SC‐N92‐4.5 V penetrate through the entire particle, the formation of block kinks subsequently leads to particle stratification, accompanied by the destruction of the crystal structure and Li⁺ loss (Figure 7c). Meanwhile, the transport pathway in the ordered RS phase undergoes a complete transformation, Li⁺ is essentially impeded at the RS interface, leading to a pronounced concentration gradient and stress concentration. Furthermore, the evolution during prolonged cycling reveals that the layered structure remains fundamentally unchanged during the initial O3–O1 phase transition, characterised by rapid Li⁺ diffusion. During the O1–LiNi₂O₄ phase transition, the crystal lattice lacks sufficient ability to accommodate Li⁺, which impedes Li⁺ diffusion and leads to a severe phase transformation characterised by lattice deformation and stress accumulation. Ultimately, this results in the gradual formation of ordered RS, accompanied by the complete destruction of the lithium‐ion transport channels and the proliferation of cracks.

FIGURE 7.

The schematic diagram during long cycles. Structural evolution of underlying reaction mechanism for (a) SC‐N92‐4.3 V, (b) SC‐N92‐4.4 V and (c) SC‐N92‐4.5 V. The Li⁺/Ni²⁺ mixed process with phase transition for (c) SC‐N92‐4.3 V and (d) SC‐N92‐4.5 V.

Understanding the phase transition mechanism of ultra‐high nickel oxides is essential for optimizing the crystal structure and fine‐tuning the test voltage. The O3–O1 phase transition process involves crystal distortion or deformation in the cathode, the transformation is characterized by the generation and accumulation of mechanical stress, with high voltage further accelerating the O3–O1 phase transition. Throughout this process, Li and Ni ions progressively replace each other, ultimately leading to a complete transformation of the lattice structure. The phase transition is a continuous process evolving from O3 (Figure S34) to O1 and subsequently to O1–LiNi₂O₄ and LiNi₂O₄–Ni₃O₄–RS, accompanied by the dissolution of TMs and ongoing alterations in the lattice structure (Figure 7d).

3. Conclusions

By integrating atomic‐resolution HAADF–STEM, MD simulations and COMSOL, we elucidated the relationship between internal phase transformations, lattice deformations, and stress generation within single‐crystal cathodes under varying cut‐off voltages at 45°C. Increasing the cut‐off voltage induces interior lattice deformation due to heterogeneous Li⁺ concentration distribution and transition metal dissolution coupled with SEI layer formation, thus promoting additional phase transformation processes (O3–RS) and influencing stress generation and accumulation. Notably, our findings reveal a novel chemo‐mechanical degradation pattern: the O3–O1, O1–LiNi₂O₄ and subsequent LiNi₂O₄–Ni₃O₄–RS phase transformations within high‐Ni layered cathodes proceed more rapidly at elevated cut‐off voltages. Contrary to previous observations, our analysis revealed not only RS phases in the crack region but also a substantial presence of LiNi₂O₄ and Ni₃O₄ phases across various micro‐crack areas. The interplay between phase transformation and stress elucidates that severe deformation, particularly during phase transformation processes, exacerbates the generation and accumulation of internal stress. Cracks emerge when localized regions of the particle cannot withstand the accumulated stress. Importantly, cracks can manifest during any bending phase transformation process, and the observed internal stress generation suggests that the predominant degradation mechanism in single‐crystal materials is driven by bending‐induced phase transformations. Elevated cut‐off voltages can induce the O3–O1 phase transition and are more prone to forming an ordered RS phase that lacks Li⁺ transport channels. Conversely, a stable crystal structure characterised by minimal phase transitions substantially enhances electrochemical performance under appropriately controlled voltage conditions. Furthermore, this novel understanding of the degradation mechanisms in layered cathodes is expected to offer valuable insights for refining testing methodologies, thus enabling the realisation of higher energy densities for practical industrial applications.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adma73901‐sup‐0001‐SuppMat.docx.

Supporting File 2: adma73901‐sup‐0002‐MovieS1‐S10.zip.

Data Availability Statement

The data that support the findings of this study are available in the Supplementary Material of this article.