Cobalt-free composite-structured cathodes with lithium-stoichiometry control for sustainable lithium-ion batteries

High-Ni NMC cathode active materials offer high capacity and lower cost but suffer from surface reconstruction, oxygen release, transition metal dissolution, bulk fatigue, and cracking, limiting practical deployment. Co in layered Li(NiMnCo)O2 facilitates structural ordering and cycling stability during calcination, yet its cost, environmental impact, and supply risk motivate eliminating Co. Co-free Ni/Mn-based layered cathodes (e.g., LiNi1MnO2) face Li/Ni disorder, cycling instability, and voltage fade due to Ni2+ introduction and magnetic interactions that exacerbate Li/Ni mixing. Conventional synthesis of high-Ni CAMs uses excess Li to ensure stoichiometry, but this promotes large particle growth and residual Li species that harm surface stability. The research question is whether controlling lithium stoichiometry during calcination can produce a Co-free LiNi0.95Mn0.05O2 with stable cycling by tailoring phase evolution and particle morphology. The study proposes Li-deficient synthesis to form a layered–rocksalt composite with small primary particles and reduced anisotropic strain, improving electrochemical performance without Co.

Prior work established Co’s role in promoting Li/Ni ordering and stability in high-Ni layered oxides, and highlighted persistent challenges in fully eliminating Co from LIB cathodes. Studies on Ni/Mn-based Co-free systems showed unavoidable Li/Ni mixing that worsens with higher Mn content, leading to capacity decay and voltage fade. Standard practice of adding excess Li during calcination ensures layered stoichiometry but often yields large cuboid particles and residual Li compounds (e.g., Li2CO3), negatively affecting cycling. Intermediate Li-containing rocksalt phases commonly form during solid-state synthesis and influence kinetic pathways. A recent report on composite LiNiO2 synthesized via controlled Li incorporation showed improved cycling versus traditional layered LiNiO2. Other approaches to enhance stability and reduce particle size rely on coatings/dopants (e.g., W, Mo, Nb), which add cost and complexity. This work builds on these insights by systematically varying Li/TM (0.90–1.10) to probe how Li stoichiometry governs phase fractions, crystallite/domain size, particle morphology, and electrochemical behavior in Co-free LiNi0.95Mn0.05O2.

Synthesis: Ni0.95Mn0.05(OH)2 precursors were produced continuously in a Taylor Vortex Reactor via hydroxide co-precipitation (NiSO4·6H2O and MnSO4·H2O, Ni:Mn=95:5, 2 M; NaOH 4 M as precipitant; NH4OH 4 M as chelating agent; pH 11.92±0.02; 52±0.2 °C; inner cylinder 800 rpm). The washed and dried hydroxides were mixed with LiOH·H2O at Li:(Ni+Mn)=0.90, 0.95, 1.00, 1.025, 1.05, or 1.10. Calcination was performed in O2 flow (1 L/min) with a two-step schedule: 600 °C for 12 h then 720 °C for 12 h. Composition was verified by ICP (ICP-AES/ICP-MS). Morphology was examined by SEM; local structure by aberration-corrected HAADF-STEM. Bulk structure and phase fractions were quantified by synchrotron XRD with Rietveld refinement; TXM-XANES 3D mapping assessed Ni valence and phase distribution. Time-resolved in situ synchrotron XRD tracked phase progression and crystallization during calcination using pellets heated to 600 °C and 720 °C. Electrochemistry: Cathodes were prepared with active material:Super P:PVDF=8:1:1 in NMP, coated on carbon-coated Al foil, loading 7–8 mg/cm2, assembled into coin cells with Li metal counter electrode and LiPF6 in EC/DEC electrolyte in Ar glovebox (O2/H2O <0.1 ppm). Cells were conditioned at 0.1 C between 2.7–4.4 V for three cycles, then cycled at 0.5 C for 100 cycles; additional 0.1 C tests assessed voltage fade. Modeling: Mesoscale simulations generated 2D precursor microstructures (~35 nm average) and simulated calcination-driven phase transformation and particle growth across Li/TM values and temperatures, capturing lithiation-induced crystallization vs liquid phase sintering regimes.

- Structure and phase: Li-deficient LiNi0.95Mn0.05O2 (Li/TM=0.95) forms a composite of a major layered phase (R3m) with a minor Li-containing rocksalt (Fm3m) phase; Rietveld refinement gives ~19.6 mol% rocksalt in NM9505-0.95Li. As Li/TM increases from 0.90 to 1.025, rocksalt fraction decreases from ~23% to ~3%; at 1.05–1.10, a single layered phase refines well. TXM-XANES shows homogeneous Ni valence distribution; HAADF-STEM confirms nanometer-scale intergrowth of layered and rocksalt within a common ccp oxygen framework. - Morphology: Li-deficient samples (Li/TM<1.0) yield small primary particles (<100 nm) with elongated rod-like shapes; near-stoichiometric and Li-excess samples produce larger cuboid particles (hundreds of nm) with broader size distributions. Excess Li leads to residual Li2CO3 on surfaces. - Calcination kinetics: In situ XRD reveals Li-containing rocksalt forms as an intermediate and transforms to layered upon heating; transformation and layered domain growth are significantly faster with Li excess (1.05Li) than with Li deficiency (0.95Li). Modeling reproduces an abrupt increase in particle growth rate for Li/TM>1, attributed to liquid phase sintering when excess Li is present. - Electrochemical performance (2.7–4.4 V): NM9505-0.95Li exhibits first-cycle CE >90% and discharge capacity of 226 mAh/g, outperforming NM9505-1.05Li (CE ~85%, 218 mAh/g). After 100 cycles at 0.5 C, capacity retention is ~90% for 0.95Li vs ~75% for 1.05Li. Voltage fade is minimal for 0.95Li (<0.05 V at 0.1 C) compared to >0.1 V for 1.05Li. dQ/dV shows reduced and broadened H2–H3 transition features for 0.95Li, indicating smoother transitions and reduced anisotropic strain. - Stability of composite: After 100 cycles, rocksalt fraction in NM9505-0.95Li remains ~17.5%, close to the pristine ~19.6%, indicating retention of the intergrown composite structure. - Composition dependence: Increasing Mn to 10% (NM9010) improves retention (~83.8% at 0.5 C, 100 cycles) but reduces initial capacity (198 mAh/g at 0.1 C; 160 mAh/g at 0.5 C), supporting low Mn as preferable for balancing capacity and stability. - Mechanism: Below stoichiometry (Li/TM≤1), lithiation-induced crystallization dominates, consuming Li into the lattice and limiting particle growth, yielding layered–rocksalt nanocomposites with low anisotropic lattice change. Above stoichiometry (Li/TM>1), excess Li persists in molten form, enabling liquid phase sintering and particle coarsening; residual Li compounds degrade surfaces.

Controlling Li stoichiometry during calcination directly addresses the challenge of stabilizing Co-free high-Ni layered cathodes. Li deficiency intentionally retains a small, uniformly distributed rocksalt fraction epitaxially intergrown with the layered phase, reducing anisotropic lattice expansion/contraction and smoothing the H2–H3 transition that typically undermines cycling stability in Ni-rich materials. The resultant sub-100 nm primary particles alleviate mechanical stress, enhance Li transport, and reduce first-cycle losses, collectively yielding higher CE, capacity, and capacity retention with negligible voltage fade compared to Li-excess materials. In contrast, Li excess accelerates layered ordering and promotes liquid phase sintering, causing large particle growth and residual surface Li species that correlate with poorer electrochemical durability. The multiscale modeling corroborates the experimentally observed transition in growth mechanism at Li/TM≈1, providing a process-structure-property framework. These results demonstrate that stoichiometry control can replace more complex coating/doping strategies to achieve structural robustness and cost efficiency in Co-free cathodes.

The study demonstrates a practical, cobalt-free LiNi0.95Mn0.05O2 cathode enabled by Li-stoichiometry control during calcination. A 5% Li-deficient route yields a layered–rocksalt nanocomposite with sub-100 nm primary particles, delivering >90% first-cycle CE, ~226 mAh/g capacity, ~90% capacity retention after 100 deep cycles to 4.4 V, and negligible voltage fade. In situ XRD and mesoscale modeling reveal that excess Li triggers liquid phase sintering and particle coarsening, whereas Li deficiency favors lithiation-induced crystallization and composite formation with suppressed anisotropic strain. The approach provides a cost-effective alternative to coating/doping strategies for stabilizing Co-free cathodes. Future work could explore broader composition spaces (e.g., varying Ni/Mn ratios), optimize Li/TM windows for different precursors and thermal profiles, scale synthesis, and validate performance in full cells and extended cycling at elevated temperatures.

- Electrochemical tests were performed in half-cells with Li metal counter electrodes; full-cell validation against practical anodes was not reported. - Cycling data span 100 deep cycles; longer-term stability and high-temperature performance require further study. - The Li-deficient composite exhibits slightly higher overpotential due to the rocksalt component’s lower Li diffusivity, which may impact high-rate performance. - Findings are demonstrated primarily for LiNi0.95Mn0.05O2; generality to other Co-free compositions and precursor morphologies needs verification. - Residual surface species (e.g., in Li-excess samples) and their mitigation strategies were identified but not optimized here.