Fe³⁺-driven tunnel engineering for stabilizing metastable ramsdellite MnO₂ in high-performance zinc-ion batteries

INTRODUCTION

Aqueous zinc-ion batteries (AZIBs) have attracted significant attention due to their high safety, low cost, and environmentally friendly characteristics^[1-3]. Among previously reported cathode materials, Mn-based oxides have become promising cathode materials for AZIBs due to their advantages in safety and high theoretical capacity. Mn-based materials from spent alkaline batteries (SABs) share chemical homology with AZIB cathodes, enabling resource regeneration from spent batteries into high-performance AZIBs^[4,5]. On the other hand, the casual discarding of SABs inevitably leads to the leakage of heavy metals such as zinc (Zn) and manganese (Mn), posing potential threats to the ecological balance and human health, thereby making the recycling and safe disposal of SABs imperative^[6].

Manganese dioxide (MnO₂) exhibits various phases in aqueous environments, including β-MnO₂, α-MnO₂, R-MnO₂ (ramsdellite-MnO₂), δ-MnO₂, λ-MnO₂, and γ-MnO₂. While β-MnO₂ demonstrates phase stability, it is limited by its (1 × 1) narrow tunnel structure; conversely, R-MnO₂ with (1 × 2) expanded tunnels significantly improves Zn²⁺ diffusion kinetics^[7-9]. Recently, our group investigated the recycling and synthesis of β-MnO₂ from SABs through hydrochloric acid hydrothermal treatment^[10]. Previous studies show that highly reactive R-MnO₂ forms initially during hydrothermal processing, but metastable R-MnO₂ transforms rapidly into β-MnO₂ with longer treatment or at high temperatures^[11]. β-MnO₂ hinders H⁺/Zn²⁺ extraction/insertion due to its narrow tunnels^[12]. In contrast, (1 × 2) tunnels in R-MnO₂ enhance ion insertion kinetics^[13]. Huang et al. synthesized an R-MnO₂-based composite material with supercapacitor characteristics and excellent electrochemical performance^[14]. Chen et al. prepared R-MnO₂ through the low-valent ion doping; the material exhibits high oxygen vacancy concentration, further improving the electrical conductivity and reactivity^[15]. Nevertheless, traditional synthesis methods face challenges in producing stable R-MnO₂, necessitating controllable and convenient strategies for its preparation^[16,17].

Meanwhile, challenges remain; lattice collapse of metastable phases (e.g., R-MnO₂) during ion insertion causes pulverization of cathode materials. Over the years, much research has employed multidimensional regulation strategies, including ionic doping, surface engineering, and defect engineering, to optimize the physicochemical properties of materials^[18]. Pan et al. synthesized Fe-MnO/C via annealing to achieve an MnO crystal structure and demonstrated that Mn-O-Fe bond formation promotes charge transfer kinetics while mitigating structural collapse, thereby synergistically improving rate capability and cycling stability^[19]. Zhong et al. showed that iron doping effectively mitigates the Jahn-Teller effect, stabilizes lattice water, and prevents lattice collapse, thus regulating the layered structure in AZIBs^[20]. Although ionic doping is widely utilized to stabilize MnO₂, the stabilization of metastable phases during synthesis remains underexplored^[21,22].

Herein, we demonstrate that Fe³⁺ doping acts as a dual-function regulator by thermodynamically stabilizing R-MnO₂ through reduced surface energy and enlarged (1 × 2) tunnels, while kinetically enhancing Zn²⁺ diffusion via optimized Mn⁴⁺/Mn³⁺ ratios that suppress Jahn-Teller effects. Through a synergistic H⁺/Fe³⁺ hydrothermal process, spent ZnMn₂O₄ waste is successfully converted into stable R-Fe_xMn_1-xO₂ nanocrystals, achieving retention of the metastable R-phase, a high surface area of 218 m² g^-1, and low charge-transfer resistance. Electrochemical performance reaches 286.8 mAh g^-1 at 0.1 A g^-1 with 88.9% capacity retention after 1000 cycles. This study establishes Fe³⁺ doping as a universal design principle for metastable functional materials, offering simultaneous advancements in AZIB cathode engineering and sustainable battery upcycling.

EXPERIMENTAL

RESULT AND DISCUSSION

R-Fe_xMn_1-xO₂ was synthesized via a pretreatment of SABs (sequential cleaning with mixed alkali solutions and acid solutions) followed by a 12 h hydrothermal reaction at 140 °C co-induced by H⁺ and Fe³⁺, while β-MnO₂ was prepared identically without the addiction of Fe³⁺ [Figure 1A]. SEM images and XRD patterns [Figure 1B-D and Supplementary Figures 1-4] reveal the overall morphology of the original SABs and Mn-based materials. β-MnO₂ exhibits a regular granular structure, while R-Fe_xMn_1-xO₂ displays irregular shapes with numerous adhered and agglomerated small particles. XRD analysis revealed that [Figure 1C] all the diffraction peaks of R-Fe_xMn_1-xO₂ correspond to R-MnO₂ (JCPDS #42-1316), demonstrating the stability of the R-MnO₂ structure upon Fe³⁺ doping. In contrast, β-MnO₂ was identified as rutile type (JCPDS #24-0735). Although R-MnO₂ is generally metastable, Fe³⁺ doping stabilized the main R-MnO₂ phase in R-Fe_xMn_1-xO₂ using this synthesis method. To validate the hydrothermal regeneration pathway and the phase transformation, comprehensive materials characterization was conducted. Combined with the material source and characterization results in Supplementary Figure 2, we confirmed the presence of impurities including Zn, C, and organics in the raw materials. During pretreatment steps, Zn and organics were effectively removed, while trace residual organics underwent carbonization during the hydrothermal process, resulting in conductive carbon that does not compromise material performance. BET surface area analysis [Figure 1E, Supplementary Figure 1C and D] demonstrated a substantial increase in specific surface area (from 78.78 m² g^-1 to 218.01 m² g^-1) and higher mesopore volume for R-Fe_xMn_1-xO₂ compared to β-MnO₂. Combined with Rietveld refinement of XRD patterns, the breaking of the long-range crystal order reduced particle size, enhanced specific surface area, and enlarged pore volume of R-Fe_xMn_1-xO₂, establishing abundant active sites for H⁺/Zn²⁺ extraction/insertion.

Figure 1. (A) Schematic illustration of the phase evolutions from SABs to β-MnO₂ and R-Fe_xMn_1-xO₂; SEM image and BET surface area of (B) β-MnO₂ and (D) R-Fe_xMn_1-xO₂ (x≈3 wt.%); (C) XRD patterns; (E) N₂ adsorption/desorption isotherms and corresponding pore size distributions (inset); (F) Fe 2p XPS spectra; (G) Surface morphology and surface quantities of the pristine surface. SEM: Scanning electron microscopy; BET: brunauer-emmett-teller; XPS: X-ray photoelectron spectrometer; SABs: spent alkaline batteries; XRD: X-ray diffraction.

High-resolution XPS analysis of the Fe 2p in R-Fe_xMn_1-xO₂ revealed a characteristic spin-orbit splitting energy of 13.9 eV between Fe 2p_1/2 and Fe 2p_3/2 peaks, confirming iron predominantly exists as Fe³⁺ [Figure 1F]^[23]. Through Fe³⁺ doping, the Mn 2p spectra exhibited two peaks at 642.28 eV and 653.68 eV, which correspond to the characteristic peaks of Mn⁴⁺ in MnO₂^[24]. Meanwhile, peak area analysis revealed R-Fe_xMn_1-xO₂ had a higher Mn⁴⁺ content (75%) than β-MnO₂ (60%), which is anticipated to effectively mitigate Jahn-Teller distortion, reduce Mn³⁺ disproportionation, and enhance conductivity [Supplementary Figures 5 and 6]. The O 1s spectra indicated greater O vacancy concentration in R-Fe_xMn_1-xO₂, facilitating rapid electron transfer. As potent defects, these oxygen vacancies significantly disrupt the lattice and inhibit crystal growth, which also contributed to enhanced nanoscale refinement of the material^[25]. DFT calculations [Figure 1G and Supplementary Figure 7] demonstrated that Fe³⁺ doping reduces surface energy (favoring the formation of R-Fe_xMn_1-xO₂), enhances the surface thermodynamic stability of R-MnO₂, significantly releases lattice stress, modifies the anisotropy of crystal growth and improves ion transport kinetics.

TEM analysis provided detailed elucidation of the material’s morphological characteristics. Figure 2A and B corroborated the morphological features observed by SEM. Compared with β-MnO₂ [Supplementary Figure 8]; the particle dimensions of R-Fe_xMn_1-xO₂ were significantly reduced, consistent with the BET surface area analysis results. SAED and High-Resolution Transmission Electron Microscopy (HRTEM) patterns verified the polycrystalline nature of R-Fe_xMn_1-xO₂. SAED exhibited well-defined diffraction spots corresponding to the characteristic R-MnO₂ (020) and (210) planes, confirming complete phase transformation to R-MnO₂ [Figure 2B₃]^[26]. The HRTEM image of R-Fe_xMn_1-xO₂ clearly shows lattice fringes of (400) (d = 0.24 nm) and (211) (d = 0.22 nm), forming an included angle of 61.8° and it can be assigned to R-MnO₂^[27]. In contrast, β-MnO₂ shows lattice fringes of 0.22 nm (200) and 0.21 nm (111) with θ = 47.7° [Figure 2A]. These results were consistent with XRD patterns and confirm the structure of R-Fe_xMn_1-xO₂, which facilitates the insertion/extraction of Zn²⁺ compared to β-MnO₂. EDS analysis [Figure 2C] demonstrated the presence of C, Mn, O and Fe elements, with no detectable Zn, confirming the formation of homogeneous Fe³⁺-doped R-Fe_xMn_1-xO₂, which regulates the lattice structure and optimizes ion diffusion.

Figure 2. The TEM and HRTEM images with corresponding SAED images of (A) β-MnO₂, (B) R-Fe_xMn_1-xO₂. (C) corresponding elemental mapping images of R-Fe_xMn_1-xO₂ powders. TEM: Transmission electron microscopy; HRTEM: high-resolution transmission electron microscopy; SAED: selective area electron diffraction.

To verify the role of Fe³⁺ doping in enhancing electrode stability and reaction kinetics, we assembled coin cells to evaluate the electrochemical performance of the materials.

We systematically evaluated the effect of Fe³⁺ feeding amount on the electrode materials and found that the optimal electrochemical performance was achieved at a Mn:Fe feeding ratio of 100:5 [Supplementary Figures 9-11]. Figure 3A shows the cycling performance of the β-MnO₂ and R-Fe_xMn_1-xO₂ electrodes. The results demonstrate enhanced cycling stability for R-Fe_xMn_1-xO₂ cathodes compared to β-MnO₂ cathodes. At 1.0 A g^-1, the R-Fe_xMn_1-xO₂ cathode exhibited specific capacity of 161.9 mAh g^-1 and achieved 88.9% capacity retention after 1000 cycles. In contrast, the β-MnO₂ cathode showed a specific capacity of 52.58 mAh g^-1 and retained only 48.8 % after cycling [Figure 3A]. The superior cycling capability of R-Fe_xMn_1-xO₂ can be attributed to Fe³⁺ doping, which optimizes ion diffusion pathways and enhances the stability of Mn-O bonds (~1.91 Å) through substitution with shorter Fe-O bonds (~1.89 Å). Additionally, the R-Fe_xMn_1-xO₂ cathodes demonstrated much better rate performance than β-MnO₂ cathodes [Figure 3B and Supplementary Figure 11], delivering specific capacities of 286.8, 269.7, 248.5, 206.8, and 163.3mAh g^-1 at current densities of 0.1, 0.2, 0.5, 1.0, and 1.5 A g^-1, respectively. The β-MnO₂ exhibited poor rate capability, achieving only 30.9 mAh g^-1 at 1.5 A g^-1. Figure 3C illustrates the galvanostatic charge-discharge curve (GCD) profiles of R-Fe_xMn_1-xO₂ and β-MnO₂ cathodes at various current densities. Notably, as current density increased, the second discharge platform below 1.3V [Figure 3C] in β-MnO₂ cathodes gradually diminished. Conversely, R-Fe_xMn_1-xO₂ maintained better plateau retention at high rates, contributing more significantly to the overall capacity.

Figure 3. Electrochemical performance of β-MnO₂ and R-Fe_xMn_1-xO₂ cathodes of (A) cycling performance [Coulombic Efficiency (CE)]; (B) rate capabilities; (C) The GCD curves; (D) GITT profiles and corresponding ion diffusion coefficients of AZIBs; (E) CV curves at scan rate of 0.1 mV s^-1. (F) radar chart comparing properties of Mn-based cathodes for AZIBs. GCD: Galvanostatic charge-discharge; GITT: galvanostatic intermittent titration technique; AZIBs: aqueous zinc-ion batteries; CV: cyclic voltammetry.

Furthermore, Galvanostatic Intermittent Titration Technique (GITT) was employed to evaluate Zn²⁺/H⁺ diffusion kinetics. The GITT curves for β-MnO₂ and R-Fe_xMn_1-xO₂ cathodes, along with the corresponding diffusion coefficients (D) were presented in Figure 3D. The D value was calculated using $$ \mathrm{D}_{\mathrm{Zn}^{2+}}=\frac{4}{\pi \iota }\left(\frac{\mathrm{n}_{\mathrm{M}}\mathrm{~V}_{\mathrm{M}}}{\mathrm{~s}}\right)^{2}\left(\frac{\Delta \mathrm{Es}}{\Delta \mathrm{Et}}\right)^{2} $$. (D: diffusion coefficients; ι: the relaxation time; nM: the number of moles; VM: molar volume of the electrode material; S: the electrode/electrolyte contact area; ΔEs: the voltage change induced by the pulse; ΔEt: the voltage change during galvanostatic charge/discharge). The R-Fe_xMn_1-xO₂ cathodes exhibited significantly higher ion diffusion coefficients throughout the entire discharge process. This kinetic advantage correlates with the stabilized (1 × 2) tunnel structure and optimized cation diffusion pathways of R-Fe_xMn_1-xO₂. Besides, CV profiles at 0.1 mV s^-1 [Figure 3E] revealed larger redox peaks for the R-Fe_xMn_1-xO₂ cathode, indicating higher electrochemical activity. More stable redox profiles demonstrated better reversibility, attributed to enhanced mechanical stability of ion transport channels by Fe³⁺ doping. The peak splitting observed in the CV curves at 1.60-1.65 V is influenced by the Zn²⁺/H⁺ co-insertion mechanism. This peak splitting originates from the energy barrier differences between H⁺ and Zn²⁺ deintercalation pathways, which leads to divergent transport kinetics and stepwise oxidation processes. In addition, the R-Fe_xMn_1-xO₂ showed smaller ΔE values between redox peaks, signifying superior electrochemical reversibility, reduced polarization, improved energy retention, and enhanced overall efficiency. Collectively, the cathode material obtained in this work exhibits superior overall performance [Figure 3F and Supplementary Table 1].

These findings suggest that Fe³⁺ doping stabilizes the R-MnO₂ structure, forming the R-Fe_xMn_1-xO₂ cathode, which not only alleviates lattice strain to enhance structural resilience against H⁺/Zn²⁺ extraction/insertion during cycling but also accelerates reaction kinetics. To figure out the structural transitions and reversibility, in situ XRD analysis was performed on the R-Fe_xMn_1-xO₂ cathode [Supplementary Figure 12]. During discharge, new diffraction peaks emerged corresponding to Zn₄SO₄(OH)₆·5H₂O (JCPDS #039-0688), attributed to precipitation involving dissolved H⁺ ions from the electrolyte. These peaks disappeared during charging. Notably, the peak intensities of R-MnO₂ remained basically unchanged throughout the cycling process, confirming the structural stability of R-Fe_xMn_1-xO₂ under electrochemical operation.

By plotting i/v^1/2 versus v^1/2 [i(V) = k₁v^1/2 + k₂v] at different potentials, the values of k₁ were obtained to calculate the percentage of capacitance contribution. The capacitive contribution for the R-Fe_xMn_1-xO₂ electrode increased gradually from 47% to 87%, while ranging between 33% and 73% for the β-MnO₂ electrode [Figure 4A and Supplementary Figure 13], demonstrating the relatively higher capacitance contribution of R-Fe_xMn_1-xO₂ cathode^[28]. Thus, the enhancement stems from the Fe³⁺-induced porous nanostructure of R-Fe_xMn_1-xO₂, which increases the specific surface area and provides additional active sites for faradic reactions, thereby improving surface pseudo-capacitance. Post-cycling XPS analysis [Figure 4B and Supplementary Figure 14] revealed higher Mn⁴⁺ content in R-Fe_xMn_1-xO₂ compared to β-MnO₂. This effectively suppresses the Mn³⁺ proportion, mitigating Jahn-Teller distortion-induced manganese dissolution and enhancing capacity retention. The O 1s spectra showed significantly intensified H-O-H peaks in cycled R-Fe_xMn_1-xO₂, indicating guest-bonded water that provides shielding by reducing positive charge density at vertex oxygen sites within the host frameworks^[29]. Based on above, Fe³⁺ provided more active sites and expedited reaction kinetics.

Figure 4. Capacitive behaviors and both electrochemical and morphological analysis after 1000 cycles of β-MnO₂ and R-Fe_xMn_1-xO₂ cathodes: (A) Diffusion-controlled and surface-controlled contributions to capacity at different scan rates; (B) Mn 2p and O 1s spectra; (C) Nyquist plots at the open circuit voltage state (Supplementary Figure 15 illustrates the electrical equivalent circuit used for fitting EIS spectra); (D and E) EIS spectra of β-MnO₂ and R-Fe_xMn_1-xO₂ cathodes during charge process; (F) SEM images of β-MnO₂ and R-Fe_xMn_1-xO₂ cathodes. EIS: Electrochemical impedance spectroscopy; SEM: scanning electron microscopy.

EIS before and after 1000 cycles (at 1 A g^-1) further elucidated the role of Fe doping in enhancing interfacial stability, as shown in Figure 4C-E. Due to abundant electrochemically active sites and accelerated ion diffusion kinetics, the charge transfer resistance (R_ct) value after cycling was significantly lower for R-Fe_xMn_1-xO₂ (562.9Ω) than for β-MnO₂ (1,720 Ω) [Supplementary Table 2]^[30]. EIS analysis during cycling stages revealed larger semicircles for both cathodes during charging and post-cycling, correlating with degradation mechanisms. However, R-Fe_xMn_1-xO₂ maintained more favorable reaction kinetics^[31,32]. The SEM images confirmed significantly more dissolution in β-MnO₂ than in R-Fe_xMn_1-xO₂, consistent with the performance degradation mechanisms in Mn-based cathodes. The dissolution of Mn-based materials reduced surface active sites, impaired ion diffusion driving force, and increased R_ct, leading to elevated impedance and irregular morphologies [Figure 4F]^[33-35]. This post-cycle analysis highlights how Fe³⁺ doping in manganese-based cathodes effectively mitigates the capacity fading inherent to manganese oxide cathodes in AZIBs.

CONCLUSIONS

Fe³⁺ doping fundamentally transforms the structural stability and electrochemical behavior of metastable ramsdellite MnO₂ for zinc-ion batteries. By introducing Fe³⁺ during the hydrothermal conversion of spent ZnMn₂O₄, we achieve R-Fe_xMn_1-xO₂ with engineered (1 × 2) tunnels, reduced surface energy, and optimized Mn⁴⁺/Mn³⁺ ratios. The Fe-mediated mechanism suppresses phase transition to β-MnO₂, alleviates lattice strain, and enhances Zn²⁺/H⁺ diffusion kinetics, validated by DFT calculations showing lowered surface energy and EIS revealing reduced charge-transfer resistance. Consequently, R-Fe_xMn_1-xO₂ delivers a high capacity of 286.8 mAh g^-1 at 0.1 A g^-1, exceptional rate capability (163.3 mAh g^-1 at 1.5 A g^-1), and long-term stability (88.9% capacity retention after 1000 cycles). Post-cycling analyses confirm Fe³⁺ mitigates Mn dissolution and structural degradation. This study highlights Fe³⁺ doping as a universal tunnel-engineering strategy to unlock the potential of metastable MnO₂ phases, while offering a sustainable route to repurpose battery waste into high-performance cathodes.