The relentless pursuit of higher energy density and lower cost in electrochemical energy storage has positioned the lithium-ion battery at the forefront of technological innovation. As demands evolve from portable electronics to large-scale electric vehicles and grid storage, the limitations of conventional cathode materials, particularly their reliance on costly and geopolitically sensitive elements like cobalt, have become increasingly apparent. This drives significant research into alternative, resource-abundant materials that can deliver superior performance without compromising sustainability or economics.
Among the promising candidates, the spinel LiNi0.5Mn1.5O4 (LNMO) stands out as a quintessential high-voltage, cobalt-free cathode. Its operation around 4.7 V vs. Li+/Li, primarily from the Ni2+/Ni4+ redox couple, unlocks a high theoretical energy density of approximately 650 Wh kg-1. This intrinsic high-voltage characteristic is a double-edged sword; while it directly contributes to high energy output, it pushes the electrochemical window far beyond the stability limit of state-of-the-art carbonate-based electrolytes. Consequently, the cathode-electrolyte interface (CEI) in a standard lithium-ion battery becomes a site of relentless parasitic activity.
The instability at this high-potential frontier manifests through several interconnected degradation pathways. First, the electrochemical oxidation of electrolyte components (e.g., ethylene carbonate, dimethyl carbonate) generates a complex mixture of organic and inorganic species. While some form a passivating layer, this naturally formed CEI on LNMO is often inhomogeneous and mechanically fragile. Second, and more critically, trace moisture in the electrolyte reacts with the LiPF6 salt to produce hydrofluoric acid (HF). HF is a potent etchant that corrodes the LNMO surface, leading to the dissolution of transition metal (TM) ions, predominantly Mn2+ and Ni2+. This dissolution triggers the infamous “crosstalk” effect, where dissolved TMs migrate through the electrolyte and deposit on the graphite anode. These deposits catalyze further decomposition of the solid electrolyte interphase (SEI) on the anode, consuming active lithium and electrolyte, leading to rapid capacity fade and impedance growth in the full lithium-ion battery cell. The overall degradation can be conceptualized by a series of parasitic reactions:
Electrolyte Oxidation: $$\text{Solvent (e.g., EC)} \rightarrow \text{Polymeric/Organic Species} + \text{CO}_2 \uparrow + \text{e}^-$$
HF Generation: $$\text{LiPF}_6 + \text{H}_2\text{O} \rightarrow \text{LiF} + \text{POF}_3 + 2\text{HF}$$
Cathode Dissolution: $$\text{LiNi}_{0.5}\text{Mn}_{1.5}\text{O}_4 + x\text{HF} \rightarrow \text{Li}_{1-x}\text{Ni}_{0.5}\text{Mn}_{1.5}\text{O}_{4-x} + x\text{LiF} + \text{Mn}^{2+} + \text{Ni}^{2+} + \frac{x}{2}\text{H}_2\text{O}$$
Anode Crosstalk: $$\text{Mn}^{2+}/\text{Ni}^{2+} + 2\text{e}^- \rightarrow \text{Mn}/\text{Ni} (\text{deposited on anode})$$ $$\text{Deposited TMs catalyze SEI growth: } \text{Li}^+ + \text{e}^- + \text{Solvent} \xrightarrow{\text{catalyst}} \text{Irreversible SEI products}$$
To combat these issues, surface engineering of the cathode material is paramount. Conventional approaches, such as atomic layer deposition (ALD) or wet-chemical coating followed by high-temperature annealing, aim to create a physical barrier. While effective, these methods often involve complex, multi-step processes and can result in rigid ceramic coatings with poor lattice matching, potentially hindering ionic transport. A more streamlined and potentially more conformal approach is the use of functional additives directly introduced into the electrode manufacturing process. These slurry additives can migrate to the active material surface during slurry mixing and coating, subsequently reacting or polymerizing to form a protective layer in situ. This method integrates seamlessly with existing lithium-ion battery manufacturing lines, offering a practical and scalable solution.
In this comprehensive study, we propose and investigate tetraethyl orthosilicate (TEOS, Si(OC2H5)4) as a novel, effective, and simple slurry additive for stabilizing LNMO cathodes. The rationale is based on the hydrolysis and condensation chemistry of TEOS. During the slurry preparation using N-methyl-2-pyrrolidone (NMP), which contains trace water, TEOS undergoes hydrolysis, followed by condensation reactions on the surface of LNMO particles. This process in situ generates an ethoxy-functionalized polysiloxane (EPS) film that intimately coats the cathode particles. This EPS film serves a dual purpose: (1) as an artificial, robust CEI with a resilient Si―O―Si network resistant to high-voltage oxidation, and (2) as an HF scavenger, where its remaining ethoxy groups react with HF to form stable Si―F bonds, thereby mitigating TM dissolution. This work provides a detailed mechanistic understanding of how this simple additive tailors the electrode-electrolyte interface, leading to dramatically improved performance in both half-cells and, more importantly, in practical graphite-based full lithium-ion battery configurations, even under elevated temperature stress.
Experimental Methodology and Material Characterization
The LNMO electrodes were prepared via a standard slurry-casting procedure. The baseline slurry consisted of 85 wt% commercial LNMO powder, 10 wt% acetylene black, and 5 wt% polyvinylidene fluoride (PVDF) binder in NMP solvent. For the modified electrodes, specific amounts of TEOS (2, 5, and 10 wt% relative to LNMO mass) were added to this mixture. The slurries were cast onto aluminum foil and dried thoroughly. The resulting electrodes are denoted as LNMO (pristine), LNMO-T0.02, LNMO-T0.05, and LNMO-T0.10.
Material characterization confirmed the successful formation of the EPS layer without altering the bulk structure of LNMO. X-ray diffraction (XRD) patterns of the electrode films showed identical peak positions and shapes, corresponding solely to the disordered spinel phase (Fd$\bar{3}$m space group). No crystalline impurities from TEOS decomposition were detected, indicating the amorphous nature of the formed polysiloxane. Fourier-transform infrared (FTIR) spectroscopy provided direct evidence of the polymer film. Powder scraped from the TEOS-modified electrodes showed distinct absorption bands at 1085 cm-1 and 1009 cm-1, attributable to Si―O―C and Si―O―Si stretching vibrations, respectively. The intensity of these peaks scaled with the amount of TEOS added.
High-resolution transmission electron microscopy (HRTEM) offered visual confirmation. While pristine LNMO particles exhibited clean surfaces, LNMO-T0.05 particles were uniformly covered by an amorphous layer with a thickness ranging from 1.5 to 5.5 nm. This variation is typical for solution-based coating processes but confirms the formation of a continuous, conformal film. Energy-dispersive X-ray spectroscopy (EDS) mapping further verified the homogeneous distribution of silicon over the LNMO particles, co-localizing with the oxygen signal from the surface film. The bulk morphology of the secondary particles remained unchanged, as observed by scanning electron microscopy (SEM).
Electrochemical Performance Evaluation
The impact of the in-situ formed EPS interface on electrochemical behavior was systematically evaluated in lithium half-cells. Initial galvanostatic charge-discharge profiles at 1C rate (1C = 147 mA g-1) were nearly identical for all electrodes, featuring the characteristic high plateau at ~4.7 V and a small shoulder near 4.0 V from the residual Mn3+/Mn4+ redox. This indicates that the thin EPS layer does not obstruct Li+ intercalation/de-intercalation. The initial discharge capacities and Coulombic efficiencies were slightly lower for TEOS-modified electrodes, suggesting a minor initial impedance from the fresh polymer film, which is typical for artificial interface layers.
The rate capability, however, revealed an advantage for the optimized composition. While all cells showed good rate performance due to the 3D Li+ diffusion channels in spinel, the LNMO-T0.05 electrode delivered the highest specific capacity at ultra-high rates (e.g., 40C). This suggests that once formed and conditioned, the EPS interface facilitates charge transfer kinetics, possibly by providing a more stable and uniform pathway for Li+ transport compared to the continuously evolving native CEI on pristine LNMO.
The most dramatic improvement was observed in long-term cycling stability. At a 2C rate, the capacity retention after 1000 cycles starkly highlighted the efficacy of the additive.
| Electrode | Initial Discharge Capacity (mAh g-1) | Discharge Capacity at Cycle 1000 (mAh g-1) | Capacity Retention (%) | Average Capacity Fade per Cycle (%) |
|---|---|---|---|---|
| LNMO (Pristine) | 119.1 | 61.3 | 51.4 | 0.0486 |
| LNMO-T0.02 | 116.1 | 82.6 | 71.1 | 0.0289 |
| LNMO-T0.05 | 117.9 | 94.9 | 84.6 | 0.0154 |
| LNMO-T0.10 | 112.5 | 90.3 | 80.3 | 0.0197 |
The LNMO-T0.05 electrode demonstrated exceptional stability, retaining 84.6% of its capacity, compared to only 51.4% for the pristine electrode. The performance of LNMO-T0.10, while good, was slightly inferior, indicating that an excessive amount of additive might lead to a thicker insulating layer, slightly hampering kinetics. The voltage polarization during cycling was significantly smaller for LNMO-T0.05, as visualized by the tighter gap between charge and discharge plateaus in later cycles compared to the rapidly widening gap in the pristine cell. The average discharge voltage decay, a critical metric for energy retention in a lithium-ion battery, was also markedly suppressed.
Electrochemical impedance spectroscopy (EIS) and galvanostatic intermittent titration technique (GITT) were employed to probe the interfacial and bulk kinetic evolution. Fitted EIS data showed that while the LNMO-T0.05 cell started with slightly higher charge-transfer resistance (Rct), its overall impedance after long-term cycling was substantially lower than that of the pristine cell. This cross-over point is crucial: the artificial EPS layer initially adds a small barrier but prevents the runaway growth of impedance associated with continuous CEI repair and TM dissolution-induced anode degradation. GITT analysis provided the apparent Li+ diffusion coefficient (DLi+). The calculated values followed the trend:
$$D_{Li^+}^{\text{cycled}} (\text{LNMO-T0.05}) > D_{Li^+}^{\text{cycled}} (\text{Pristine LNMO}) > D_{Li^+}^{\text{fresh}} (\text{LNMO-T0.05}) \approx D_{Li^+}^{\text{fresh}} (\text{Pristine LNMO})$$
This confirms that the primary benefit of the EPS film is not in enhancing intrinsic bulk diffusion but in preserving the interface and bulk structure over time, leading to superior sustained kinetics in a working lithium-ion battery.
Full-Cell Validation and the Suppression of Crosstalk
The ultimate test for any high-voltage cathode modification is its performance in a practical full lithium-ion battery, where limited lithium inventory amplifies the impact of any side reaction. We constructed full cells pairing the LNMO-based cathodes with commercial graphite anodes (N/P ratio ~1.1). The results unequivocally validated the strategy.
The LNMO-T0.05||graphite full cell delivered outstanding cycling stability, retaining 83.3% of its initial capacity after 500 cycles at 2C, whereas the pristine LNMO||graphite cell retained only 53.4%. This represents a transformative improvement in cycle life. Furthermore, the cells were tested at an elevated temperature of 50 °C, a condition that accelerates all parasitic reactions. Under this harsh environment, the LNMO-T0.05||graphite cell maintained 63.8% capacity after 200 cycles, dramatically outperforming the pristine cell, which faded to 26.7% retention. This highlights the robustness of the EPS-derived interface in stabilizing the high-voltage lithium-ion battery even under thermal stress.
Post-mortem analysis was conducted to uncover the mechanistic roots of this enhancement. Inductively coupled plasma mass spectrometry (ICP-MS) of cycled lithium metal anodes (from half-cells) quantified the extent of TM dissolution. The results are summarized below:
| Electrode | Cycles | Mn Deposition (μg) | Ni Deposition (μg) | Relative Mn Dissolution (% of Pristine) | Relative Ni Dissolution (% of Pristine) |
|---|---|---|---|---|---|
| LNMO | 200 | 2.87 | 0.95 | 100% | 100% |
| LNMO-T0.05 | 200 | 2.02 | 0.58 | 70.3% | 61.1% |
| LNMO | 1000 | 5.41 | 2.54 | 100% | 100% |
| LNMO-T0.05 | 1000 | 4.32 | 1.36 | 79.8% | 53.7% |
The data clearly shows that the EPS film significantly suppresses the dissolution of both Mn and Ni, with Ni dissolution being inhibited more effectively. This reduction in TM crosstalk is a primary reason for the enhanced full-cell performance. X-ray photoelectron spectroscopy (XPS) of cycled electrodes provided chemical-state evidence. On cycled LNMO-T0.05, the F 1s spectrum showed a pronounced peak at ~685.5 eV, which can be assigned to both metal fluorides (M-F) and, critically, silicon fluoride (Si-F) species. The concurrent presence of Si-F and Si-O bonds in the Si 2p spectrum confirmed that the EPS film actively scavenges HF via the reaction:
$$\text{Si―OC}_2\text{H}_5 + \text{HF} \rightarrow \text{Si―F} + \text{C}_2\text{H}_5\text{OH}$$
Furthermore, the Mn 2p spectrum from the cycled LNMO-T0.05 cathode showed a lower relative concentration of Mn2+ species compared to the pristine cycled cathode. Since Mn2+ is the dissolution-prone state and is often found in CEI components like MnF2, its reduction indicates less Mn dissolution and redeposition. SEM images of electrodes after long cycling revealed that pristine LNMO secondary particles suffered from severe microcracking, likely due to gas evolution from electrolyte oxidation and internal stress. In contrast, the LNMO-T0.05 particles maintained their structural integrity, indicating a much less corrosive interfacial environment.
Discussion: A Unified Mechanism for Interface Tailoring
The integration of electrochemical data and post-mortem analysis allows us to construct a coherent model for how the TEOS slurry additive tailors the interface and stabilizes the high-voltage lithium-ion battery. The process and its benefits can be summarized in a staged mechanism:
Stage 1: In-Situ Film Formation during Electrode Fabrication. During slurry mixing and drying, TEOS hydrolyzes and condenses on the LNMO particle surface: $$\text{Si(OC}_2\text{H}_5)_4 + 2\text{H}_2\text{O} \rightarrow \text{SiO}_2 + 4\text{C}_2\text{H}_5\text{OH}$$ (Simplified; forms a polysiloxane network with residual ethoxy groups). This forms a conformal, amorphous EPS film with a robust Si―O―Si backbone and terminal -OC2H5 groups.
Stage 2: Electrochemical Activation and HF Scavenging. Upon initial charging in the lithium-ion battery, the high potential and trace HF initiate two key processes. First, the film densifies and integrates into the evolving CEI. Second, and more importantly, the ethoxy groups act as sacrificial sites, reacting with corrosive HF to form stable, insoluble Si-F bonds within the film matrix: $$\equiv\text{Si―OR} + \text{HF} \rightarrow \equiv\text{Si―F} + \text{ROH}$$. This dramatically lowers the concentration of free HF at the cathode surface.
Stage 3: Long-Term Stabilization. With HF neutralized, the primary drivers for TM dissolution are curtailed. The stable Si-O-Si/Si-F network acts as a durable artificial CEI, preventing direct contact between the aggressive high-voltage cathode surface and the electrolyte. This leads to:
- Reduced CEI Growth: Less continuous electrolyte oxidation and decomposition.
- Mitigated Cathode Degradation: Minimal TM dissolution and lattice oxygen loss.
- Elimination of Destructive Crosstalk: Reduced TM migration to the anode, preserving anode SEI and active lithium.
- Preserved Morphology: Suppression of particle cracking from gas and stress.
The net effect is a significant reduction in the overall cell impedance growth rate and active material loss. The Li+ diffusion kinetics are preserved over time because the interface remains stable and the bulk cathode structure is protected. The improvement can be quantitatively linked to the suppression of the main degradation factors. If we model capacity fade as a sum of contributions from Li loss (e.g., at anode SEI, $k_{SEI}$), cathode active material loss ($k_{CAM}$), and impedance rise ($k_{R}$), the additive’s role is to minimize all three coefficients:
$$\frac{dQ}{dN} = -(k_{SEI} + k_{CAM} + k_{R})$$
For the modified electrode: $k_{SEI, mod} \ll k_{SEI, pristine}$ (less crosstalk), $k_{CAM, mod} < k_{CAM, pristine}$ (less dissolution/surface degradation), $k_{R, mod} < k_{R, pristine}$ (stable interface).
Conclusion and Perspectives
This work demonstrates that tetraethyl orthosilicate (TEOS), employed as a simple slurry additive, is a highly effective strategy for in-situ engineering of a stable cathode-electrolyte interface on high-voltage LiNi0.5Mn1.5O4 cathodes. The additive undergoes hydrolysis-condensation during electrode processing to form a conformal, ethoxy-functionalized polysiloxane (EPS) film. This multifunctional film serves as a robust artificial CEI and an active HF scavenger, addressing the two core instability issues plaguing high-voltage lithium-ion batteries: electrolyte decomposition and transition metal dissolution.
The optimized electrode delivered exceptional cycling stability, retaining 84.6% capacity after 1000 cycles in half-cells and, more critically, 83.3% after 500 cycles in practical graphite-based full cells. The strategy also proved effective under elevated temperature (50 °C) conditions, underscoring its robustness. The mechanism, elucidated through a suite of electrochemical and spectroscopic techniques, confirms the suppression of detrimental crosstalk and the preservation of interfacial kinetics.
This approach offers distinct advantages: it is simple, scalable, and compatible with standard lithium-ion battery manufacturing processes, avoiding the complexity of ex-situ coating methods. It opens a new avenue for interface design through slurry additives. Future work could explore a library of other alkoxysilane (e.g., with different alkyl chains, or functional groups like amine or epoxy) or other hydrolyzable precursors to tailor the film’s properties—such as its elasticity, ionic conductivity, or affinity for specific anions/cations—further optimizing performance for next-generation lithium-ion batteries. Furthermore, investigating its applicability to other high-voltage or high-capacity cathode systems (e.g., Li-rich layered oxides, Ni-rich NCM) could broaden its impact on the development of advanced energy storage technologies.