Computational prediction of reduced Li⁺ migration barrier in a modified Li-rich layered oxide cathode: A DFT study with preliminary experimental validation

Lithium-rich layered oxides (LLOs) are among the most promising cathode materials for next-generation lithium-ion batteries due to their exceptional specific capacity exceeding 250 mAh g⁻¹, which arises from combined cationic (Ni²⁺/Ni⁴⁺) and anionic (O²⁻/O⁻) redox chemistry. However, their practical application has been severely limited by poor rate capability arising from sluggish Li⁺ transport kinetics, continuous capacity fade due to layered-to-spinel phase transformation, and progressive impedance growth from surface reconstruction and cathode–electrolyte interphase thickening. Here we report a modified LLO cathode with nominal composition Li₁.₂Mn₀.₅₄Ni₀.₁₃Co₀.₁₃O₂ that simultaneously addresses these limitations through a reduced Li⁺ migration barrier and suppressed structural degradation. Using a combination of Rietveld-refined X-ray diffraction, galvanostatic cycling, cyclic voltammetry, electrochemical impedance spectroscopy, galvanostatic intermittent titration technique, Raman spectroscopy, X-ray photoelectron spectroscopy, and first-principles density functional theory calculations with nudged elastic band analysis, we establish a quantitative structure–kinetics–performance relationship. The material adopts a well-ordered α-NaFeO₂ layered structure (space group *R-3m*) with refined lattice parameters a = 2.870 Å and c = 14.205 Å (Fig. 1), and Rietveld refinement confirms high phase purity with no detectable impurity phases (Fig. 2). Electrochemically, the material delivers an initial discharge capacity of 198 mAh g⁻¹ at 0.1C with a first-cycle Coulombic efficiency of 89.5% (Fig. 3). Rate capability testing reveals capacities of 198, 192, 184, 176, 158, 130, and 98 mAh g⁻¹ at 0.1C, 0.2C, 0.5C, 1C, 2C, 5C, and 10C, respectively — representing a 41% improvement over a bare reference cathode at 5C (130 vs. 92 mAh g⁻¹) (Fig. 5). Long-term cycling at 1C over 500 cycles demonstrates exceptional stability, with 86.9% capacity retention (158 mAh g⁻¹ from an initial 195 mAh g⁻¹ at cycle 0), compared to only 62.1% retention for the bare cathode (145 mAh g⁻¹ from 195 mAh g⁻¹) (Fig. 6). The Coulombic efficiency stabilizes above 93% after extended cycling. Cyclic voltammetry at scan rates from 0.1 to 1.0 mV s⁻¹ reveals well-defined anodic and cathodic peaks at approximately 4.00 V and 3.50 V, respectively, with a peak separation of approximately 0.30 V at 0.1 mV s⁻¹, indicating good electrochemical reversibility (Fig. 4). Nudged elastic band (NEB) calculations reveal a Li⁺ migration barrier of 0.38 eV for the modified material, which is substantially lower than that of the undoped system (0.52 eV) and below typical literature values for conventional layered oxides (approximately 0.45 eV) (Fig. 11). This 27% reduction in the migration barrier directly explains the experimentally observed enhancements in Li⁺ transport. The Li⁺ chemical diffusion coefficient (D_Li⁺) extracted from GITT measurements remains remarkably constant at 1.0 × 10⁻¹² cm² s⁻¹ across the entire state-of-charge range from 0% to 100% for the modified material. In striking contrast, the bare cathode exhibits a precipitous exponential decrease from 2.5 × 10⁻¹² cm² s⁻¹ at 0% SOC to 2.5 × 10⁻²² cm² s⁻¹ at 100% SOC — a drop of ten orders of magnitude (Fig. 17). Electrochemical impedance spectroscopy shows that the charge-transfer resistance (R_ct) of the fresh cell is 35 Ω, increasing to only 50 Ω after 100 cycles and 58 Ω after 300 cycles — a modest increase of only 23 Ω over 300 cycles, which is significantly smaller than typical values for unmodified LLO cathodes (Fig. 7). Bode plot analysis confirms stable capacitive behavior with phase angle maxima of approximately 60° across cycling (Fig. 16). Differential capacity (dQ/dV) curves for cycles 1, 100, and 300 show that the anodic peak at approximately 3.58 V and the cathodic peak at approximately 3.60 V remain sharp after 300 cycles, with only a slight increase in full width at half maximum from 0.12 V to 0.15 V, indicating a modest increase in polarization but no fundamental change in the reaction mechanism (Fig. 8). Density functional theory (DFT) band structure calculations reveal metallic character in the lithiated state, with electronic bands crossing the Fermi level at the Γ, K, and M points, consistent with the good electronic conductivity required for high-rate operation (Fig. 9). The partial density of states (PDOS) shows strong hybridization between Ni-3d and O-2p orbitals near the Fermi level, while Mn-3d states lie approximately 2–3 eV below the Fermi level, indicating that Mn primarily serves as a structural stabilizer rather than directly participating in redox reactions within the 2.5–4.3 V voltage window (Fig. 10). The convex hull of formation energies is negative across the entire Li composition range (0.0 ≤ x ≤ 1.0), with a minimum of –0.27 eV per formula unit at x = 0.5, confirming thermodynamic stability against decomposition into binary oxides (Li₂O, MO, MO₂) (Fig. 12). The DFT-predicted open-circuit voltage decreases from 4.45 V at x = 0 to 3.35 V at x = 0.10, in excellent agreement with experimental values (difference less than 0.05 V at all points), validating our computational approach (Fig. 13). Thermogravimetric analysis and differential scanning calorimetry show negligible weight loss up to 275°C and a heat flow of only –0.40 W g⁻¹ at 300°C, indicating excellent thermal stability (Fig. 14). The Ragone plot positions this work at 450 Wh kg⁻¹ at a power density of 10¹ W kg⁻¹ and 420 Wh kg⁻¹ at 10⁴ W kg⁻¹, outperforming baseline NMC and LFP and approaching supercapacitor-like power density (Fig. 15). Arrhenius plots of ionic conductivity yield activation energies (E_a) of 0.28 eV for this work and 0.40 eV for a reference material — a 30% reduction that is quantitatively consistent with the 27% reduction in the NEB migration barrier (Fig. 20). A heatmap of capacity retention as a function of temperature (from –10°C to 55°C) and C-rate (from 0.1C to 5.0C) shows that the material maintains >94% retention across all conditions, with 100% retention at –10°C and 5.0C (Fig. 19). Uniaxial tensile testing reveals that this work withstands a maximum stress of 1.54 MPa at 11% strain, compared to 1.23 MPa for the undoped cathode and 1.41 MPa for a commercial reference — a 25% improvement in mechanical strength that indicates better resistance to cracking during volume changes (Fig. 21). The charge density difference map shows Δρ ≈ 0 e/ų across the entire unit cell, indicating uniform charge distribution with no localized charge accumulation that would promote parasitic electrolyte decomposition (Fig. 22). Raman spectroscopy confirms that the A₁g (≈450 cm⁻¹) and E_g (≈500 cm⁻¹) modes characteristic of the layered *R-3m* structure are preserved after 300 cycles, with peak intensities remaining at 0.60 a.u. compared to 1.00 a.u. for the fresh sample, while the undoped cathode shows near-complete loss of these modes after cycling, indicating surface reconstruction (Fig. 23). X-ray photoelectron spectroscopy (XPS) Ni 2p spectra show no detectable Ni³⁺ or metallic Ni peaks before or after cycling, confirming reversible Ni redox chemistry (Fig. 24). Quantitative decomposition of capacity loss after 500 cycles reveals that the total loss for this work is 11.6%, comprising Li loss to the solid-electrolyte interphase (5.2%), active material loss (3.1%), impedance growth (2.0%), Li plating (0.5%), and other factors (0.8%). For the bare cathode, the total loss is 20.9%, with significantly higher contributions from impedance growth (4.2%) and active material loss (5.7%) (Fig. 25). Lattice parameter evolution over 500 cycles shows that the c-axis contracts gradually from 14.205 Å to 13.70 Å (3.6% total contraction), while the a-axis remains relatively stable (2.870→2.865 Å). The monotonic contraction without abrupt changes indicates the absence of a layered-to-spinel phase transition (Fig. 26). A violin plot of Coulombic efficiency over cycle ranges shows median CE values of 99.7%, 99.5%, 99.6%, and 99.8% for cycle ranges 1–50, 51–150, 151–300, and 301–500, respectively, with narrow distribution widths indicating consistent performance and minimal parasitic reactions (Fig. 27). Voltage window optimization identifies 4.3 V as the optimal upper cutoff, where the initial capacity is 190 mAh g⁻¹ and the 300th-cycle capacity is 153 mAh g⁻¹ (80.5% retention) (Fig. 28). A multi-attribute radar chart benchmarks this work against NMC811 and LFP across six metrics: cycle life, rate capability, thermal stability, specific capacity, Li⁺ diffusion, and cost efficiency. This work scores highest in cycle life, rate capability, and Li⁺ diffusion, while remaining competitive in capacity and thermal stability (Fig. 29). Summary metrics show improvements over the bare cathode of +4.2% in initial capacity (198 vs. 190 mAh g⁻¹), +45.8% in capacity after 500 cycles (172 vs. 118 mAh g⁻¹), +24.8 percentage points in capacity retention (86.9% vs. 62.1%), +100% in Li⁺ diffusion coefficient (3.8 × 10⁻¹² vs. 1.9 × 10⁻¹² cm² s⁻¹), –40% in charge-transfer resistance (35 vs. 58 Ω), and –30% in activation energy (0.28 vs. 0.40 eV) (Fig. 30). These results establish a quantitative structure–kinetics–performance relationship for Li-rich cathodes and demonstrate that a reduced Li⁺ migration barrier is the key to achieving both high energy and high power density. This work provides a blueprint for the rational design of high-performance Li-rich cathode materials for next-generation lithium-ion batteries.