The ever-increasing demand for high-energy-density portable electronics and electric vehicles has led to the massive proliferation of lithium-ion batteries (LIBs). Consequently, a correspondingly vast stream of spent LIBs is being generated, posing significant environmental risks and resource sustainability challenges. Spent LIBs contain valuable and often scarce metals like lithium and cobalt, alongside hazardous electrolytes, making their improper disposal a serious concern. Efficient recycling is therefore imperative, not only to mitigate environmental pollution but also to reclaim critical materials and support a circular economy for battery technologies. Among various cathode chemistries, lithium cobalt oxide (LiCoO2, LCO) has been a workhorse material for decades due to its high volumetric energy density and stable voltage profile. Despite the rise of newer materials, LCO still holds a substantial share of the LIB market, particularly in consumer electronics, leading to a continuous and significant generation of spent LCO cathodes. Current recycling strategies often involve hydrometallurgical processes for metal extraction or direct regeneration of the cathode material. However, these methods can be complex, generate secondary waste, or face issues with batch consistency due to the variable state-of-health of spent batteries.
Parallel to the recycling challenge, the quest for sustainable and cost-effective alternatives to LIBs for large-scale energy storage has intensified. The sodium-ion battery (SIB) has emerged as a prime candidate, leveraging the abundance and low cost of sodium. The operating principles of a sodium-ion battery are analogous to those of LIBs, but they offer potential advantages in safety and resource availability. Developing high-performance cathode materials is central to advancing sodium-ion battery technology. Layered transition metal oxides, specifically the P2-type structure (with prismatic Na coordination and two alkali layers per unit cell), are promising cathodes for sodium-ion batteries. In particular, P2-Na0.67Fe0.5Mn0.5O2 (NFMO) is attractive due to its high theoretical capacity, the natural abundance and low toxicity of Fe and Mn, and its competitive energy density. However, the practical application of this material in sodium-ion batteries is hindered by issues such as irreversible phase transitions during cycling, Jahn-Teller distortion associated with Mn3+, and electrolyte degradation at high voltages, all contributing to rapid capacity fading.
To address these limitations, cationic doping has proven to be an effective strategy. Doping with elements like Al, Ti, V, or Mg can enhance structural stability, suppress phase transitions, and improve Na+ diffusion kinetics in P2-type cathodes for sodium-ion batteries. This work proposes a novel, integrated approach that tackles both the recycling challenge and the performance enhancement need. We demonstrate the direct utilization of critical elements (Li and Co) recovered from spent LCO cathodes as dopants to modify and improve the electrochemical performance of NFMO cathodes for sodium-ion batteries. This strategy offers a simple, low-cost, and efficient route for waste valorization while simultaneously engineering a better cathode material for next-generation sodium-ion batteries.
Experimental Synthesis and Characterization
Spent mobile phone batteries were fully discharged and disassembled in an argon-filled glovebox. The recovered LCO cathode sheets were washed with dimethyl carbonate (DMC) to remove residual electrolyte. The active material was then scraped off the aluminum current collector and calcined at 800 °C in air for 10 hours to remove carbon additives and the polyvinylidene fluoride (PVDF) binder, yielding purified waste LCO powder.
The modified NFMO cathodes, with a nominal composition of Na0.67–1.33xLixCox(Fe0.5Mn0.5)1–2xO2, were synthesized via a sol-gel method. Calculated amounts of the waste LCO powder (corresponding to x = 0, 0.035, 0.070, 0.105, 0.140) were first dissolved in a diluted nitric acid solution with a small amount of H2O2 at 60 °C under stirring. To this solution, stoichiometric amounts of NaNO3 (5% excess to compensate for Na loss), Mn(CH3COO)2·4H2O, and Fe(NO3)3·9H2O were added. Citric acid, with a molar ratio of 1:1 to the total metal ions (Na+Fe+Mn), was introduced as a chelating agent. The mixture was heated under continuous stirring to evaporate water and form a viscous gel. The gel was dried at 120 °C for 12 hours to obtain a porous precursor, which was then ground and subjected to a two-step calcination: 450 °C for 5 hours to decompose organic species, followed by 900 °C for 15 hours in air to form the final crystalline product. The pristine NFMO (x=0) was synthesized identically for comparison.
The phase purity and crystal structure were analyzed by X-ray diffraction (XRD). Morphology and elemental distribution were examined using scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). The chemical states of surface elements were probed by X-ray photoelectron spectroscopy (XPS). For electrochemical evaluation, cathode slurries were prepared by mixing active material, Super P carbon, and PVDF binder (8:1:1 by weight) in N-methyl-2-pyrrolidone (NMP) and coated onto aluminum foil. CR2032 coin cells were assembled in an Ar glovebox using the prepared cathode, sodium metal anode, glass fiber separator, and an electrolyte of 1 M NaClO4 in a mixture of ethylene carbonate (EC) and propylene carbonate (PC) (1:1 by volume) with 5 wt% fluoroethylene carbonate (FEC) additive. Galvanostatic charge/discharge tests were conducted between 1.5 and 4.2 V (vs. Na+/Na). Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements were performed on an electrochemical workstation.
Structural and Morphological Evolution
The XRD patterns of all synthesized materials, both pristine and LCO-modified, confirmed the formation of a pure P2-type phase with the P63/mmc space group. No secondary phases related to Li-Co-O or other impurities were detected, indicating that Li+ and Co3+ ions were successfully incorporated into the NFMO lattice, forming a homogeneous solid solution. A closer look at the (002) peak revealed a systematic shift to higher angles with increasing LCO-derived dopant content (x).
This peak shift is a direct consequence of lattice contraction. Since the ionic radii of the dopants are smaller than those of the host ions they replace (r(Li+) ≈ 0.76 Å, r(Co3+) ≈ 0.545 Å vs. r(Fe3+) ≈ 0.645 Å, r(Mn3+) ≈ 0.650 Å, r(Mn4+) ≈ 0.530 Å), their substitution leads to a decrease in unit cell parameters. Rietveld refinement was employed to quantify these changes. The results are summarized in Table 1.
| Composition (x) | a (Å) | c (Å) | Volume (Å3) |
|---|---|---|---|
| 0 (Pristine NFMO) | 2.937 | 11.285 | 84.31 |
| 0.035 (5% LCO) | 2.905 | 11.248 | 82.85 |
| 0.070 (10% LCO) | 2.903 | 11.225 | 81.90 |
| 0.105 (15% LCO) | 2.882 | 11.278 | 80.72 |
| 0.140 (20% LCO) | 2.885 | 11.179 | 80.53 |
The lattice parameter a, related to the metal-metal distance within the transition metal layer, decreased monotonically, confirming the substitution of larger Fe/Mn ions with smaller Li/Co ions. The change in the c parameter, related to the interlayer spacing, was more complex but the overall cell volume decreased consistently. This lattice contraction can influence Na+ diffusion pathways and structural stability during cycling in a sodium-ion battery.
XPS analysis provided insights into the oxidation state changes induced by doping. For the pristine NFMO, Mn existed in a mixed Mn3+/Mn4+ state, and Fe in a Fe3+/Fe4+ state. Upon Li/Co co-doping, the ratio of Mn4+ increased significantly. This is a critical finding for sodium-ion battery cathodes. The oxidation is driven by charge compensation: when low-valent Li+ (formally +1) substitutes for higher-valent transition metals (typically +3), or when Na+ vacancies are created, the average oxidation state of the remaining transition metals must increase to maintain electroneutrality. The increased Mn4+ content suppresses the Jahn-Teller distortion inherent to Mn3+ (d4 configuration), which is a major source of structural instability in these layered oxides. The XPS also confirmed the presence of Co in a Co3+/Co4+ mixed state within the doped samples.
SEM images showed that all materials exhibited a similar plate-like morphology with primary particle sizes of 3–5 μm. The doping process did not drastically alter the particle shape or size distribution. EDS elemental mapping for a representative doped sample (x=0.105) confirmed the homogeneous distribution of Na, Fe, Mn, and Co throughout the particles, further evidencing the formation of a uniform solid-solution cathode for the sodium-ion battery.
Electrochemical Performance in Sodium-Ion Batteries
The galvanostatic charge/discharge profiles of all cathodes within a sodium-ion battery configuration exhibited characteristic plateaus. The pristine NFMO showed an initial discharge capacity of 182 mAh g–1 at 0.1C. The LCO-modified samples exhibited lower initial capacities: 173, 159, 154, and 129 mAh g–1 for x = 0.035, 0.070, 0.105, and 0.140, respectively. This capacity reduction is intrinsic to the doping mechanism. In the nominal composition Na0.67–1.33xLixCox(Fe0.5Mn0.5)1–2xO2, Li+ and Co3+ primarily occupy transition metal (TM) sites. Since Li+ is electrochemically inactive in this voltage range and serves mainly as a stabilizer, and Co’s redox activity (Co3+/Co4+) occurs at a potential similar to Fe, the number of available redox-active TM ions per formula unit decreases. Furthermore, the creation of Na+ vacancies (due to the 1.33x coefficient) reduces the number of extractable Na+ ions. The theoretical capacity (Cth) can be approximated by considering the number of reversible Na+ ions (nNa) and the average redox electron transfer per formula unit. A simplified relation is:
$$C_{th} = \frac{n_{Na} \times F}{3.6 \times M_W}$$
where F is Faraday’s constant and MW is the molecular weight. As x increases, nNa decreases, leading to lower Cth.
While the specific capacity decreased, the cycling stability of the sodium-ion battery improved dramatically with Li/Co doping. After 150 cycles at 1C, the capacity retention rates were 16%, 45%, 59%, 67%, and 75% for x = 0, 0.035, 0.070, 0.105, and 0.140, respectively. The enhanced stability originates from the structural stabilization effects: 1) Suppression of the Jahn-Teller effect via increased Mn4+ content, and 2) mitigation of large volume changes due to the lower absolute capacity being delivered per cycle.
Cyclic voltammetry curves aligned with the galvanostatic data, showing redox couples corresponding to Mn3+/Mn4+ (~2.0/2.5 V) and Fe3+/Fe4+ (~3.5/4.0 V vs. Na+/Na). In the doped samples, an additional redox pair from Co3+/Co4+ was observed overlapping with the Fe redox region. With higher doping levels (x ≥ 0.105), the intensity of the Mn redox peaks diminished, indicating less participation of Mn in the redox process, which correlates with the improved cycle life.
EIS analysis provided further evidence for the improved interfacial stability in the doped cathodes. The Nyquist plots consisted of a semicircle in the high-medium frequency region (associated with charge-transfer resistance, Rct) and a sloping line in the low-frequency region (related to Na+ solid-state diffusion). The fitted Rct values are summarized in Table 2. The sample with x = 0.105 showed a moderate initial Rct which significantly decreased after 100 cycles, suggesting the formation of a more stable and conductive cathode-electrolyte interphase (CEI) during cycling. In contrast, the pristine NFMO and heavily doped (x=0.140) samples showed higher or less stable interfacial resistance.
| Composition (x) | Rct – Initial (Ω) | Rct – After 100 cycles (Ω) |
|---|---|---|
| 0 (Pristine NFMO) | ~600 | ~500 |
| 0.035 (5% LCO) | ~450 | ~400 |
| 0.070 (10% LCO) | ~700 | ~550 |
| 0.105 (15% LCO) | ~650 | ~250 |
| 0.140 (20% LCO) | ~800 | ~600 |
The rate capability of the sodium-ion battery cathodes was also assessed. As shown in Table 3, while the pristine NFMO capacity plummeted at high rates (only 8 mAh g–1 at 5C), the doped samples, particularly the one with x = 0.105, maintained significantly higher capacities. This improved rate performance is attributed to the lower polarization and more stable Rct observed in the optimally doped material, facilitating faster Na+ (de)intercalation kinetics.
| Composition (x) | 0.1C | 1C | 2C | 5C | Return to 0.1C |
|---|---|---|---|---|---|
| 0 (Pristine) | 182 | 110 | 45 | 8 | 165 |
| 0.035 | 173 | 125 | 75 | 22 | 155 |
| 0.070 | 159 | 115 | 80 | 25 | 145 |
| 0.105 | 154 | 120 | 90 | 30 | 148 |
| 0.140 | 129 | 105 | 70 | 20 | 120 |
Mechanistic Insights and Optimal Composition
The electrochemical data clearly indicates a trade-off between specific capacity and cycling stability in this doping system for sodium-ion batteries. The optimal performance was achieved with the composition Na0.53Li0.105Co0.105(Fe0.5Mn0.5)0.79O2 (x = 0.105). This material balanced several key factors:
1. Jahn-Teller Suppression: The doping introduced sufficient Mn4+ to minimize the detrimental distortion during discharge, enhancing structural reversibility.
2. Interfacial Stability: This composition promoted the formation of a robust and conductive CEI, as evidenced by the lowest cycled Rct, reducing continuous electrolyte decomposition and impedance growth.
3. Kinetic Optimization: It maintained favorable Na+ diffusion pathways despite lattice contraction, yielding the best rate performance among the series.
4. Capacity Trade-off: The sacrifice in initial capacity (~15% less than pristine) was acceptable relative to the >4-fold improvement in cycle life retention.
For higher doping levels (x = 0.140), while cycle life was good, the excessive reduction in redox-active sites and possible Li+ migration to Na+ layers likely increased impedance and degraded rate capability, making it sub-optimal. The relationship between doping level (x) and a composite performance metric (P) considering both capacity (C) and retention (R) can be conceptually described by a parabolic function, where an optimum exists:
$$P(x) = C(x) \times R(x)$$
with C(x) decreasing and R(x) increasing with x, leading to a maximum P at an intermediate x value (≈0.105 in this study).
Conclusion
This study successfully demonstrates a sustainable and effective strategy for valorizing spent lithium-ion battery cathodes by direct application in engineering high-performance cathodes for sodium-ion batteries. Critical Li and Co elements recovered from waste LiCoO2 were incorporated into the P2-Na0.67Fe0.5Mn0.5O2 lattice via a simple sol-gel process. The Li+/Co3+ co-doping induced lattice contraction, increased the Mn4+ content to suppress Jahn-Teller distortion, and fostered a more stable electrode-electrolyte interface. Although the specific capacity experienced a predictable decrease, the cycling stability and rate capability of the modified sodium-ion battery cathode were substantially enhanced. The optimal composition, Na0.53Li0.105Co0.105(Fe0.5Mn0.5)0.79O2, delivered a capacity retention of 67% after 150 cycles at 1C and a respectable capacity of 30 mAh g–1 at a high rate of 5C, significantly outperforming the pristine material. This work presents a compelling “waste-to-resource” pathway, addressing two critical challenges in energy storage simultaneously: reducing environmental impact from battery waste and advancing the development of durable, cost-effective cathode materials for the promising sodium-ion battery technology.