Graphene-Stabilized δ-MnO₂ Cathode for High-Capacity Aqueous Aluminum-Ion Batteries

Abstract

Aluminum-ion batteries (AIBs) are emerging as promising alternatives to lithium-ion batteries due to their cost-effectiveness and resource abundance. However, their practical application is hindered by low capacity, poor cycle life, and limited rate capability. In this study, δ-MnO₂ and δ-MnO₂/Graphene composite cathodes are developed and tested in aqueous aluminum-ion batteries (AAIBs) using a mixture of 0.5 M Al₂(SO₄)₃ and 0.4 M MnSO₄ as the electrolyte. The electrochemical properties are evaluated alongside treated aluminum (TAl) and Zn–Al alloy anodes. Among the configurations tested, the δ-MnO₂/Graphene|0.5 M Al₂(SO₄)₃, 0.4 M MnSO₄|Zn–Al system showed the best performance, achieving a high discharge voltage of 1.63 V, a specific capacity of 746 mAh g⁻¹, and excellent cycling stability over 352 cycles. The stabilizing effect of graphene, due to increased oxygen vacancies and the formation of Mn–O–C bonds, enhances electron and ion transport, thereby improving cathode integrity and the overall performance of the AAIB. Additionally, the Zn–Al alloy anode extends the battery’s cycle life compared to the TAl anode. This work demonstrates the potential for low-cost, high-performance AAIBs, paving the way for more sustainable and scalable energy storage solutions.

1. Introduction

AIBs have garnered significant interest as viable substitutes for LIBs due to the abundance of aluminum, low cost, and high theoretical capacity (2980 mAh g⁻¹) [1,2]. By leveraging the trivalent redox chemistry of aluminum, AIBs offer a capacity that far surpasses lithium-ion counterparts, establishing them as a prominent candidate for future energy storage innovations. AAIBs further enhance this potential by providing improved safety, environmental compatibility, and economic feasibility, making them particularly suitable for extensive applications. Unlike systems that rely on expensive ionic liquids, which primarily provide monovalent anions such as AlCl₄⁻ and Al₂Cl₇⁻, AAIBs harness the trivalent Al³⁺ ions, offering the potential for higher capacity and significantly improved overall cost-efficiency [3].

Despite these advantages, the practical realization of AAIBs faces several critical challenges. The presence of a passivating layer of aluminum oxide on the anode is a major problem because it prevents effective electrochemical reactions. Moreover, the hydrogen evolution side reaction, which happens before the aluminum deposition potential (−1.66 V vs. SHE) at the aluminum anode [4], reduces the overall efficiency and stability of the battery system. Another key limitation is the scarcity of high-performance cathode materials. A variety of approaches have been explored to tackle the issues associated with AAIBs. On the anode side, surface modification of aluminum [5,6,7,8,9] and the incorporation of aluminum alloys [1,3,8,10,11,12] have shown potential in improving anode compatibility with aqueous electrolytes, thereby mitigating corrosion and passivation issues. On the cathode side, different oxide materials such as vanadium oxides [13,14,15,16] (e.g., V₂O₅, FeVO₄, VOPO₄), bismuth oxide [17] (e.g., Bi₂O₃), and tungsten oxide [18] (e.g., WO₃) have been reported in AAIBs. Among them, manganese oxide-based cathodes [5,19,20] have demonstrated auspicious performance due to their mixed charge storage mechanisms, combining Al³⁺ intercalation and capacitive surface reactions; their structural tunability through doping or nano structuring; and the formation of stable surface layers during cycling, which can enhance electrochemical reversibility [21]. Although manganese-based oxides have shown promising performance in AAIBs [1,5,6,8,11,12,22,23], their poor cycling stability in energy systems remains a challenge due to the dissolution of manganese species into the aqueous electrolyte [24,25,26,27], necessitating effective stabilization strategies. Stabilizing MnO₂ by inhibiting the Jahn-Teller effect [28], creating oxygen vacancy (V_O) [22,29,30], and using dopants [11] can further enhance its electrochemical performance. Another practical approach involves integrating cathode materials with carbonaceous materials, such as graphene, to overcome their inherently low conductivity and structural instability during cycling [31,32,33]. The graphene not only enhances electron transport but also prevents structural degradation and increases the cathode’s surface area, all of which contribute to the improved performance in manganese oxide-based energy systems [34,35,36,37,38,39,40,41]. In our previous work [42], we demonstrated that coupling MnO₂ with graphene effectively enhanced electrochemical performance in LIBs. Although the benefits are apparent, the MnO₂/graphene cathode has not yet been studied in AAIBs, suggesting that it is a promising avenue for further AAIB development.

Closing this gap, this study proposes a system comprising a stabilized manganese oxide/graphene composite cathode, a zinc-aluminum alloy anode, and a cost-effective fluorine-free aqueous electrolyte consisting of 0.5 M Al₂(SO₄)₃ and 0.4 M MnSO₄. To the best of our knowledge, this is the first AAIB to use an MnO₂-graphene cathode. The configuration aims to enhance cycling stability and energy density, thereby supporting the development of high-performance AAIBs for practical, real-world energy storage applications.

2. Materials and Methods

2.1. Synthesis of Cathode Materials

The δ-MnO₂ and δ-MnO₂/Graphene were synthesized using the method described in our previous work [42]. A 4 mL aqueous solution containing 0.300 g Mn(NO₃)₂·4H₂O (≥99.9%) was mixed with an 8 mL aqueous solution of 0.190 g NaOH (≥99.0%) and 0.530 g of H₂O₂ (30 wt.%) under stirring for 20 min. Subsequently, the resulting precipitate was mixed with 30 mL of 2 M NaOH (all chemicals sourced from Merck, Germany). The mixture was heated for 16 h at 150 °C in a Teflon-lined autoclave [43]. After cooling, the product was collected by centrifugation, thoroughly washed with deionized water until a neutral pH was achieved, and subsequently dried under ambient conditions.

For δ-MnO₂/Graphene synthesis, 20.0 mg graphene (purity >98%, United Nanotech Innovations Pvt Ltd., India) was first ultrasonically dispersed in water, and 0.300 g Mn(NO₃)₂·4H₂O was added. The graphene–Mn-containing solution was then added to 8 mL of an aqueous solution containing 0.190 g NaOH and 1.080 g H₂O₂. After stirring for 20 min, the resulting mixture was again combined with 30 mL of 2 M NaOH. The final mixture was transferred to the autoclave and subjected to hydrothermal treatment at 150 °C for 2 h. The product was collected, washed, and dried following the same procedure.

2.2. Battery Assembly

Aluminum foils (≥99.3%, 16 μm thick) were surface-treated with an ionic liquid (IL) in an argon-filled glovebox. The foils were submerged in AlCl₃:[EMIm]Cl (1.5:1) for ten days. Before assembling the batteries, the treated foils were wiped clean with tissue paper inside the glovebox. The aqueous electrolyte was prepared by dissolving Al₂(SO₄)₃·xH₂O (x = 14–18, ≥97%, Thermo Fisher Scientific, USA) and MnSO₄.4H₂O (99%, Thermo Fisher Scientific, USA) in deionized water. δ-MnO₂ and δ-MnO₂/Graphene cathodes were fabricated by mixing the active material, carbon black (KS6L, TIMCAL, Switzerland), and polyvinylidene fluoride (PVDF, Solef-5130, Solvay, Belgium) in a weight ratio of 7:1.5:1.5 with N-Methyl-2-Pyrrolidone (NMP). After applying the prepared slurry to a 0.2 mm thick piece of 316L stainless-steel foil, it was dried for 16 h at 80 °C in an electric oven. The battery components, consisting of the prepared cathode, TAl or zinc foil (≥ 99.95 percent, 0.24 mm thick—used as a substrate to form Zn-Al anode in situ in the battery), and a glass microfiber filter (Whatman 1822-025, Cytiva, UK) as the separator, were assembled in Swagelok cells with stainless-steel current collector rods (12 mm diameter) under ambient air condition. To prevent corrosion, the rod side surfaces were completely insulated, exposing only the cross-section for electrode contact.

2.3. Electrochemical Measurements

A VMP3 multichannel potentiostat (BioLogic, France) was used for all electrochemical measurements. Galvanostatic charge/discharge and cycling experiments were conducted at 180 mA g⁻¹ with charge capacity regulation instead of a voltage limit to prevent probable side reactions and due to the lack of an explicit voltage cutoff. Each measurement sequence began with a discharge step. For anode voltage investigation, a three-electrode system with a Swagelock cell was employed, comprising a Hg/Hg₂SO₄ reference electrode, the fabricated working electrodes, and either aluminum or zinc foil as the counter electrode.

Full cells and a three-electrode setup were used for cyclic voltammetry (CV) tests. A platinum counter electrode, a glassy carbon working electrode with a 3 mm diameter (BASi, UK), and a reference electrode of Hg/Hg₂SO₄ were employed in the three-electrode setup. CV scans were conducted for full cells between 0.1–1.5 V for TAL anode batteries and 0.4–1.9 V for Zn–Al anode batteries at various scan rates. For the three-electrode configuration, the CV scans were recorded in the range of −0.9 to 0.5 V vs. Hg/Hg₂SO₄ for TAL and −1.0 to 0.5 V vs. Hg/Hg₂SO₄ for the Zn–Al system. Unless otherwise specified, all electrochemical measurements were conducted using an electrolyte composed of 0.5 M Al₂(SO₄)₃ and 0.4 M MnSO₄.

Open circuit voltage (OCV) conditions were used to perform electrochemical impedance spectroscopy (EIS) measurements, at a state of charge, using a 5 mV voltage amplitude, six data points per decade, and two averages per frequency, in the frequency range of 200 kHz to 0.1 Hz. EC-Lab software (version 11.50) was used to fit equivalent circuits.

2.4. Material Characterization

Atomic absorption spectroscopy (AAS) using a ContrAA 800F spectrometer (Analytik Jena, Germany) was employed to measure Mn and Na contents in the electrolyte after discharge. Measurements were performed at wavelengths of 403.0755 nm (Mn) and 330.237 nm (Na), with atomization using acetylene (55 L/h) and compressed air (75 L/h). For analysis, the battery components were disassembled, and the electrolyte was washed with deionized water to a total volume of 50 mL. To examine the surface topography and roughness of the cathode electrode (δ-MnO₂) before and after use in the aluminum battery, an atomic force microscopy (AFM) device (Bruker Co.,USA) was utilized.

A WITec alpha300 apyron Raman microscope (WITec, Germany) equipped with a 532 nm argon-ion laser was used to perform the Raman analysis. A Philips PW3373/00 Cu LFF DK 433781 diffractometer (Philips, Netherlands) operating at 30 mA and 40 kV over a 2θ range of 10–80° was used to perform X-ray diffraction (XRD) analysis of synthesized powders. The XRD grazing incident mode was applied to analyze the electrode surface with an incident angle (θ) fixed at 3, step size 0.02 degrees, and step time 1 s per step using a Siemens D5000 diffractometer (Siemens, Germany). Field emission scanning electron microscopy (FE-SEM) was performed using an FEI ESEM Qanta 200 instrument (FEI, USA). In addition, energy-dispersive X-ray spectroscopy (EDS, Thermo Fisher Scientific, USA) and high-resolution SEM (Hitachi S-4800, Japan) were employed. A field emission transmission electron microscope (TEM, Philips Tecnai-F20, Netherlands) was used to take the TEM image. The surface chemical composition was examined using X-ray photoelectron spectroscopy (XPS, XR50M, SPECS Surface Nano Analysis GmbH, Germany), which used a monochromatic AlKα source running at 300 W and an ultra-high vacuum chamber with a base pressure of 2 × 10⁻⁸ mbar. CASAXPS software (version 2.3.22PR1.0) was utilized to perform curve fitting.

3. Results and Discussion

3.1. Synthesized Cathode Materials Characterization

To investigate the effects of compositing δ-MnO₂ with graphene, various characterization techniques, XRD, Raman, SEM, TEM, and XPS, were used and will be discussed in detail. The XRD patterns of the synthesized manganese oxide and the graphene-composite material verify the crystalline phase of birnessite (Figure 1a), which is consistent with PDF No. 43-1456. A distinctive diffraction peak at 12.4°, corresponding to the (001) plane with a d-spacing of 7.14 Å along the c-axis, is observed in both samples. Furthermore, a subtle peak at 26.3°, as shown in the inset of Figure 1a, corresponds to the (002) plane of graphene within the δ-MnO₂/Graphene composite. The SEM images (Figure 1c) reveal the layered morphology of δ-MnO₂ in both samples. The δ-MnO₂ nanosheets in the δ-MnO₂/Graphene composite are evenly spaced throughout the graphene sheets, suggesting that δ-MnO₂ has successfully anchored onto the graphene matrix. The δ-MnO₂/Graphene morphology is shown in the TEM images of Figure 1d, which validates the distribution of nanosheet δ-MnO₂ over the graphene matrix.

Birnessite crystallite size, as determined by the Debye-Scherrer equation, was 20 nm for δ-MnO₂/Graphene and 36 nm for δ-MnO₂. The birnessite crystallites in δ-MnO₂/Graphene are therefore considerably smaller than those in δ-MnO₂.

The Raman spectra of graphene, δ-MnO₂, and the δ-MnO₂/Graphene composite material are shown in Figure 1b. In both the δ-MnO₂ and δ-MnO₂/Graphene samples, a stretching vibration of Mn–O is observed at 633 cm⁻¹. This vibration was ascribed to vibrations in the basal plane of MnO₆ octahedra within the MnO₂ crystalline structure [44]. The D-band, which is linked to structural disorder, and the G-band, which is related to the C–C stretching mode in sp²-hybridized carbon, are the two distinct peaks detected in the graphene’s Raman spectrum at 1337 cm⁻¹ and 1565 cm⁻¹, respectively. For determining the defect density in carbon materials, one commonly used metric is the intensity ratio of the D to G band (I_D/I_G). When the I_D/I_G increases from 0.75 in pure graphene to 0.94 in the manganese oxide-graphene composite, it signifies a notable rise in defect density, which may help the interfacial electron transfer between MnO₂ and graphene [45], consistent with the improved electrochemical properties of δ-MnO₂/Graphene. Furthermore, the D and G bands observed shift to 1343 cm⁻¹ and 1576 cm⁻¹, respectively, suggesting that graphene and MnO₂ are strongly interacting and transferring charges [46]. Compared to graphene, the G-band in the composite is weaker and broader, indicating an increase in defects in graphene, possibly due to edge effects or reduction during the hydrothermal synthesis process [47]. This analysis further confirms the successful composite formation and the interaction of graphene sheets with MnO₂ without the formation of by-products.

Figure 2 presents the XPS analysis comparing δ-MnO₂ and δ-MnO₂/Graphene samples. In Figure 2a, the survey XPS spectrum for the δ-MnO₂ reveals the presence of Mn, O, C, and Na peaks, confirming the absence of extraneous elements and the purity of the sample. Similarly, the survey spectrum for the δ-MnO₂/Graphene composite displays peaks corresponding to Mn 2p, O 1s, C 1s, and Na 1s, with a higher carbon content attributed to the addition of graphene.

Figure 2b–d, which demonstrates high-resolution spectra of individual elements, shows that Mn is present in 2+, 3+, and 4+ oxidation states in the Mn 2p spectrum for both δ-MnO₂ and the δ-MnO₂/Graphene composite (Figure 2b). The satellite peak (~645.5 eV), which is frequently linked to manganese oxides [48], was isolated from the Mn 2p_3/2 peaks in order to quantify the relative amounts of Mn²⁺, Mn³⁺, and Mn⁴⁺ that are claimed to exist for birnessite in the literature [49,50]. Both δ-MnO₂ and δ-MnO₂/Graphene have almost the same relative percentage of manganese oxidation states. However, the XPS O 1s spectra (Figure 2c) reveal notable structural changes. The peaks at 530.1, 531.2, and 533.4 eV, corresponding to lattice oxygen (Mn–O–Mn), V_O [38,51], and C–O or H–O–H bonds [52], respectively. In the δ-MnO₂/Graphene composite, the Mn–O–Mn peak intensity decreases, while the C–O peak intensity and V_O increase. The lower intensity of the Mn–O vibration in the Raman spectra for δ-MnO₂/Graphene could also be an additional confirmation of an increase in V_O [53]. Mn 2p_3/2 Mn 2p_1/2 exhibits a slight but steady shift in binding energy (~0.1 eV) to lower values, which supports this structural change. These changes have been ascribed to interfacial electronic interactions in composite systems and defect generation, specifically V_O, in earlier research [54,55]. It has been observed that V_O improves the mobility of charge carriers by altering electronic characteristics, which in turn speeds up the kinetics of ion insertion (EIS-verified; the following sections will address this). Moreover, the presence of V_O at the electrolyte-electrode interface influences the surface’s thermodynamic characteristics, which facilitates phase transformations [29,53]. Thus, a contributing factor to the composite’s superior cycling stability is its higher V_O content (8.8% to 17.2%). The following sections will illustrate this enhanced performance.

The C–C/C=C, C–O–Mn, C=O, O–C=O, and π-π* transitions are responsible for the peaks at 284.6, 286.1, 288, 289.6, and 291.2 eV in the C1s spectrum for the δ-MnO₂/Graphene composite [56,57]. Adventitious carbon is attributed to the peak for δ-MnO₂ at 284.6 eV, which is most likely the result of air exposure (Figure 2d). The presence of Mn–O–C bonds in δ-MnO₂/graphene indicates the formation of strong interfacial interactions between graphene and δ-MnO₂, which can help mitigate the Jahn–Teller effect [57]. This effect, associated with Mn³⁺, originates from the uneven occupancy of e_g orbitals, leading to structural distortions that lower the system energy [28]. During battery cycling, manganese dioxide is prone to partial dissolution into the electrolyte, as confirmed by AAS measurement (discussed in the section that follows). The formation of Mn–O–C bonds could therefore be an efficient way of stabilizing the MnO₂ cathode.

It is reported that the interfacial bonding in metal oxide–graphene composites can result in band gap narrowing. Ti–O–C bond formation, for instance, has been demonstrated to modify the band structure and to considerably narrow the band gap in a TiO₂-graphene system [58]. Building on our earlier findings [42], the band gap decreased from 1.7 eV in pristine MnO₂ to 1.4 eV in MnO₂/Graphene. The observed band gap reduction is probably caused by similar effects induced by the formation of Mn–O–C bonds between MnO₂ and graphene. Prior research has also documented that in MnO₂-graphene composites, this type of interfacial bonding can directly affect the band structure and reduce the energy needed for electronic transitions [59]. On the other hand, the MnO₂ nanostructures are securely anchored to the conductive graphene sheets by Mn–O–C bonds that form at the MnO₂/graphene interface [60]. therefore, while reducing mechanical strain during repeated cycling, these strong interfacial bonds enable effective electron transport between MnO₂ and graphene. So, the formation of Mn–O–C bonds, as another stabilizing effect of graphene, could enhance the structural stability of the composite and improve its cycling performance.

3.2. Study of the Battery System Utilizing TAl Anode

To investigate the effect of surface modification of Al anode on battery performance, batteries were assembled using both untreated Al and TAl anodes. As shown in Figure S1, the surface treatment of the aluminum anode enhances electrochemical performance in the TAl|δ-MnO₂ system, increasing the capacity from 255 mAh g⁻¹ to 890 mAh g⁻¹ and reducing polarization. The anode potential also shifted from −1.0 V to −1.1 V vs. Hg/Hg₂SO₄ reference electrode, while the OCV of the fresh cell (measured as the initial potential between working and counter electrode) increased by 0.15 V from 1.2 V to 1.35 V. These improvements stem from changes in anode surface chemistry that facilitate charge transfer. This supports earlier findings that an artificial solid electrolyte interphase (ASEI) layer can improve interfacial charge transfer and inhibit additional oxide formation by corroding the native Al₂O₃ film [61]. Although the cathode remained unchanged, the higher capacity with the TAl anode suggests that the untreated Al surface previously limited MnO₂ utilization due to poor interfacial kinetics. Thus, the anode plays a critical role in determining the practical capacity of the cell, even when the cathode is nominally the limiting component.

AAS analysis of the electrolyte after the first discharge of the δ-MnO₂ cathode in 0.5 M Al₂(SO₄)₃ electrolyte revealed the dissolution of manganese (1.16 mg/L) and sodium (4.48 mg/L), which can degrade battery performance. To mitigate this, MnSO₄ was added to the electrolyte to stabilize MnO₂ and preserve the cathode material. Figure S2 compares the cycling performance of cells using δ-MnO₂ and δ-MnO₂/Graphene cathodes with TAl anode under different MnSO₄ concentrations. For the δ-MnO₂ cathode, the addition of 0.2 M MnSO₄ resulted in an increase in capacity compared to the MnSO₄-free electrolyte, while 0.4 M MnSO₄ led to even higher capacity over extended cycling. A similar trend was observed for the δ-MnO₂/Graphene cathode, where the 0.4 M MnSO₄ electrolyte consistently outperformed the electrolyte without MnSO₄.

Furthermore, charge–discharge profiles (Figure S1) demonstrate that the presence of 0.4 M MnSO₄ reduces polarization and increases the discharge capacity, supporting previous findings [6] that Mn²⁺ ions contribute to Mn-oxide-based cathode stabilization in AAIBs. In addition, Mn²⁺ may also participate in redox processes, providing an auxiliary charge storage mechanism beyond Al³⁺ intercalation, as reported in the literature [25].

Based on these findings, the electrolyte composition of 0.5 M Al₂(SO₄)₃ and 0.4 M MnSO₄ was selected as the optimized composition for subsequent electrochemical evaluations.

The XRD results for the δ-MnO₂ cathode (Figure 3a) indicate that the distinct birnessite (001) diffraction peak at 12.4° disappears completely in the discharged state and does not reappear upon charging. This suggests that δ-MnO₂ undergoes a structural transformation during the first discharge. It has been reported that during this process, a new amorphous manganese oxide compound containing aluminum (AlₓMn_(1−x)O₂) forms [6]. Similarly, XRD analysis of the δ-MnO₂/Graphene cathode after the second discharge (Figure 3a) shows the absence of the birnessite peak, reinforcing the conclusion that structural change of MnO₂ occurs in both types of cathodes during the first cycle.

Charge-discharge curves (Figures S1 and S3) and CV analysis (Figure S4) for aluminum batteries using δ-MnO₂ and δ-MnO₂/Graphene cathodes further support these findings. The significant differences between the first and second cycles indicate that the aluminum is irreversibly incorporated into the cathode, thus changing the active substance. This implies again that the manganese dioxide in both cathodes undergoes an irreversible structural change during the first cycle.

To further investigate morphological changes, SEM analysis was performed on δ-MnO₂ and δ-MnO₂/Graphene electrodes before battery assembly and after the second discharge (Figure 3b). The fresh electrodes exhibit a well-defined layered morphology of manganese dioxide. However, after the second discharge, the layered morphology for the δ-MnO₂ cathode appears less coherent, and a noticeable surface change suggests structural rearrangements that could impact electrode performance. AFM analysis (Figure 3c) reveals an increase in surface roughness (R_a) after the second discharge for the δ-MnO₂ cathode, likely due to the formation of the amorphous AlₓMn_(1−x)O₂ phase and other cycling-induced structural changes. These results align with previous studies that associate increased roughness with structural modifications and void formation caused by repeated charge-discharge cycles [62].

Despite these transformations, SEM images in Figure 3b confirm that the overall layered morphology remains preserved after the second discharge, as also observed in the AFM data (Figure 3c). Notably, Figure 3b shows that in δ-MnO₂/Graphene electrodes, graphene plays a crucial role in maintaining cathode integrity. Although the newly formed cathode material became XRD-amorphous, graphene helped stabilize the manganese oxide layers, likely due to Mn–O–C interactions at the δ-MnO₂/graphene interface. This stabilizing effect, as previously discussed in the XPS section, contributes to improved cathode integrity during cycling. Consequently, the incorporation of graphene enhanced stability, electrochemical performance, and energy storage capacity.

EDS analysis provided additional insight into compositional changes in the cathode. Figure 4a verifies that the δ-MnO₂/Graphene electrode contains Mn, Na, and O prior to use, consistent with the Na-birnessite elements. The EDS mapping results further evidence a uniform elemental distribution. Signals from Fe, Ni, Cr, Mo, and Si correspond to the stainless-steel substrate. After the first charge, EDS analysis of the δ-MnO₂/Graphene cathode (Figure 4b) reveals the appearance of an aluminum peak, indicating that during the initial discharge, aluminum from the electrolyte is integrated into the cathode material. Notably, aluminum remains in the cathode after charging, supporting the irreversible formation of AlₓMn_(1−x)O₂. The absence of a sodium peak suggests a structural change in the birnessite structure, consistent with the XRD findings. The uniform distribution of the elements Al, O, C, and Mn is further revealed by the EDS map.

Figure 5a displays the second cycle charge-discharge curves for δ-MnO₂ and δ-MnO₂/Graphene electrodes at a current density of 180 mA g⁻¹. The battery incorporating the δ-MnO₂/Graphene cathode exhibits lower voltage polarization and an elevated discharge voltage compared to the δ-MnO₂ cathode. This enhancement is ascribed to improved reaction kinetics made possible by graphene [63], as confirmed by the EIS analysis (discussed later). The discharge curves reveal two distinct voltage plateaus for the δ-MnO₂ electrode at approximately 1.1 V and 0.5 V. In contrast, the δ-MnO₂/Graphene electrode exhibits higher discharge voltage plateaus at 1.3 V and 0.7 V, indicating improved energy efficiency and reduced internal resistance. With a specific capacity of 1017 mAh g⁻¹, the δ-MnO₂/Graphene electrode outperforms the pure δ-MnO₂ electrode by 171 mAh g⁻¹ (890 mAh g⁻¹). This enhancement highlights the role of graphene in forming a conductive network that mitigates polarization and improves electrochemical performance [64].

The cycling stability and coulombic efficiency of δ-MnO₂ and δ-MnO₂/Graphene electrodes are presented in Figure 5b,c. Following 45 cycles at 180 mA g⁻¹, δ-MnO₂/Graphene’s discharge capacity dropped from 934 mAh g⁻¹ to 739 mAh g⁻¹. At the same time, δ-MnO₂ underwent a significant degradation, going from 705 mAh g⁻¹ to 259 mAh g⁻¹. The greater coulombic efficiency of δ-MnO₂/Graphene compared to δ-MnO₂ is shown in Figure 5c. The rate capability test at various charge/discharge rates (Figure 5d) confirms superior rate performance for δ-MnO₂/Graphene compared to δ-MnO₂. After cycling at 3.6 A g⁻¹, the δ-MnO₂/Graphene composite exhibits a 64% capacity retention, which is significantly higher than that of pristine δ-MnO₂, which retains only 47%.

The Nyquist plots in Figure 5e, along with the fitted EIS parameters in

Table 1. Parameters of the equivalent circuit obtained from EIS data shown in Figure 5e, analyzed with EC-Lab software.

Cathode	R1 (Ω)	R2 (Ω)	R3 (Ω)	Q1 (μF.s1−n)	Q2 (μF.s1−n)
δ-MnO2	2.60 ± 0.05	36.9 ± 2.6	1333 ± 30	150 ± 5	14.5 ± 3.1
δ-MnO2/Graphene	1.78 ± 0.04	5.90 ± 0.48	1110 ± 20	530 ± 1	280 ± 4

, highlight the distinct electrochemical impedance characteristics of δ-MnO₂ and δ-MnO₂/Graphene in full cells. According to the corresponding equivalent circuit, R₁ stands for the electrolyte resistance and R₂ for the SEI layer resistance. Q₁ and Q₂ are constant phase elements (CPEs) that account for non-ideal capacitive behavior, incorporating the effects of electrode surface roughness. R₃ refers to the charge transfer resistance (R_ct) [65].

The lower ohmic resistance (R₁ = 1.78$\mathsf{\Omega }$) of δ-MnO₂/Graphene compared to δ-MnO₂ (R₁ = 2.6 $\mathsf{\Omega }$) indicates that graphene enhances electronic conductivity within the electrode, reduces resistive losses at the electrode-electrolyte interface, and improves overall electron transport pathways. Furthermore, the charge transfer resistance for δ-MnO₂/Graphene (R₃ = 1110$\mathsf{\Omega }$) is considerably lower than that observed for δ-MnO₂ (R₃ = 1333 $\mathsf{\Omega }$), demonstrating improved electrochemical reaction kinetics [41]. Another notable feature, presented in Table 1, is the significantly higher Q₂ value of the δ-MnO₂/Graphene electrode. Given the similar CPE exponent (n ≈ 0.45) in both cases, this can be interpreted as an improvement in pseudocapacitive behavior. This improvement may stem from the increased surface area and the greater number of electroactive MnO₂ sites uniformly dispersed on the graphene framework. The presence of graphene effectively minimizes MnO₂ agglomeration, promoting better exposure of active material and facilitating more efficient ion and electron transport. These effects lead to higher charge storage capacity and improved electrochemical kinetics, which contribute to the superior rate performance and overall energy storage capability of the δ-MnO₂/Graphene composite. This trend is further supported by Figure S5, where the δ-MnO₂/Graphene electrode exhibits a noticeably larger enclosed area in the CV curves compared to δ-MnO₂ alone.

The stabilizing effect of graphene, due to the increase in V_O and Mn–O–C bonds formation as discussed in the XPS section, further enhances the electrochemical performance of the δ-MnO₂/Graphene composite cathode. In the case of AAIBs, Gu et al. [22] demonstrated that introducing V_O into γ-MnO₂ significantly accelerated ion intercalation by expanding the MnO₆ octahedral framework, reducing diffusion barriers, and increasing reaction kinetics. This modification led to a lowered band gap of MnO₂ and a higher discharge capacity. Similarly, Han et al. [53] examined β-MnO₂ in Zn-ion batteries and, based on density functional theory (DFT), showed that oxygen vacancies increase ion-transport channels and enhance electronic conductivity, thereby yielding higher specific capacity and improved cycling performance. In our case, the reduced band gap and enhanced electrochemical performance of the composite are therefore likely due to the dual effect of V_O and Mn–O–C bond formation.

3.3. Study of the Battery System Utilizing Zn–Al Anode

The results in the previous section indicated that the TAl anode in aqueous electrolyte exhibits limited cycling stability, enduring only 45 charge/discharge cycles. This instability is mainly due to the progressive degradation of the aluminum surface, where the initially beneficial ASEI gradually loses effectiveness due to passivation. In aqueous electrolytes, these ASEIs tend to degrade during cycling, resulting in unstable byproducts [66]. Despite the improvement in initial cycle performance, it has been reported that the ASEI does not support the electrochemical reduction of Al³⁺ to metallic aluminum, but instead the development of corrosion and hydrogen gas contributes to secondary reactions [61]. As shown in Figure S7a, the TAl anodes extracted from the batteries after cycling exhibit aluminum dissolution over time, leading to an irreversible loss of active material due to self-corrosion. In the SEM image shown in Figure S7b, the EDS results reveal the formation of a non-conductive surface layer containing aluminum, oxygen, and sulfur on the aluminum surface. Additionally, the SEM image in Figure S7c clearly displays pits and pores on the Al surface, indicating aluminum dissolution due to self-corrosion and the potential occurrence of gas evolution. These findings are in line with previous literature reports: under highly acidic (pH < 4) or alkaline (pH > 8.6) conditions, aluminum corrodes to Al³⁺ or AlO₂⁻ with a subsequent development of hydrogen gas until either the metal or water is exhausted, whereas under medium pH conditions, an oxide film may be formed on the surface [67]. Thus, the low cycling stability may be attributed to anode passivation, anode dissolution, or depletion of the electrolyte.

To mitigate these issues, alloying aluminum with other metals has been explored as a strategy to suppress self-corrosion, prevent hydrogen evolution reactions, and inhibit the formation of inactive oxide layers [68]. Here, zinc has been chosen as a substrate for the formation of the Zn-Al anode material by co-deposition of Zn²⁺ and Al³⁺ on the Zn. The Zn–Al|δ-MnO₂ system exhibited an initial OCV of 1.65 V, with the anode potential at −1.4 V vs. Hg/Hg₂SO₄. In contrast, the TAl|δ-MnO₂ system showed an initial OCV of 1.35 V. XRD analysis (Figure 6a) was conducted on both bare zinc and the anode after the third charge in a battery with δ-MnO₂/Graphene cathode. The results indicate that all detected peaks correspond to metallic zinc (PDF-40-0831), with no evidence of metallic aluminum or aluminum oxide formation. The significant variation in peak intensity ratios, particularly I₍₀₀₂₎/I₍₁₀₀₎, provides further evidence for the occurrence of aluminum alloying with zinc [8]. These two peaks also exhibit a subtle shift of approximately 0.02° toward lower angles, while the positions of the other peaks remain unchanged.

SEM and EDS analyses were used to examine the surface morphology and elemental distribution of the anode (Figure 6b–d). Compared to the initially smooth morphology of the zinc substrate (Figure 6a), the roughened surface observed after the third charge in the SEM image (Figure 6c) indicates the formation of a coating layer on the zinc after battery operation. EDS analysis (Figure 6d) confirms a uniform distribution of aluminum, with an atomic concentration of 2% in the alloy. The co-existence of zinc and aluminum in the formed layer further validates the formation of the Zn-Al alloy.

Figure 6e–g present the SEM image and EDS analysis of the δ-MnO₂/Graphene cathode after the third charge in a battery with a Zn-Al alloy anode. Zinc signals are absent from the EDS results, indicating that Zn²⁺ ions are not involved in the cathodic discharge process, hence offering no contribution to the battery’s capacity. However, Zn²⁺ is involved in the stripping/plating process at the anode and plays a role in forming the alloy with aluminum, as also reported in previous studies [8,69]. The appearance of aluminum peaks in the EDS indicates the incorporation of aluminum into the cathode structure, resembling behavior observed in batteries with an aluminum anode. The uniform distribution of Al, O, and Mn elements and the absence of sodium peaks strongly imply that birnessite has undergone a structural change and the formation of a new phase, most likely AlₓMn_(1−x)O₂.

In a battery consisting of a Zn-Al anode, Figure 7a compares the charge/discharge profiles of δ-MnO₂ and δ-MnO₂/Graphene cathodes. The δ-MnO₂/Graphene cathode demonstrates a higher discharge voltage and lower polarization compared to δ-MnO₂. Furthermore, the capacity of the second cycle for δ-MnO₂/Graphene is 584 mAh g⁻¹, which is 248 mAh g⁻¹ more than that of δ-MnO₂.

The coulombic efficiency and cycling performance of δ-MnO₂ and δ-MnO₂/Graphene electrodes are shown in Figure 7c,b. As shown in Figure 7b, δ-MnO₂/Graphene retains a capacity of 746 mAh g⁻¹ after 352 cycles at 180 mAh g⁻¹, while δ-MnO₂ maintains a capacity of 642 mAh g⁻¹ after 107 cycles. Both δ-MnO₂ and δ-MnO₂/Graphene electrodes exhibit a gradual increase in capacity over the initial cycles (Figure 7a,b), indicating an activation process. This behavior is probably due to the progressive formation of an AlₓMnO₂ phase, as reported by Pan et al. [23], which enhances Al³⁺ storage capability. Additionally, Ejigu et al. [1] attributed similar capacity increases to gradual surface activation via electrodeposition of Mn species from Mn²⁺ additive. Based on charge/discharge profile evolution over cycling, they proposed enhanced intercalation/extraction kinetics of H⁺ and Al³⁺ ions, as well as improved Al plating/stripping behavior at the anode. Additionally, Figure 7c demonstrates that, similar to the capacity, the coulombic efficiency of both electrodes improves over cycles, approaching nearly 100% in later stages. These findings align with our observations. Supporting this, the Nyquist plot also shows the emergence of a Warburg diffusion tail after 100 cycles, further indicating enhanced ion transport and activation of the cathode material over time, as will be discussed later in this section.

The rate capability test at various charge/discharge rates (Figure 7d) highlights the superior rate capability of δ-MnO₂/Graphene over δ-MnO₂, with higher capacity retention even after cycling under high-rate conditions up to 1.8 A g⁻¹. At this current, δ-MnO₂/Graphene delivers 444 mAh g⁻¹, whereas δ-MnO₂ achieves 334 mAh g⁻¹. Furthermore, after prolonged cycling at 1.8 A g⁻¹, δ-MnO₂/Graphene retains 95% of its initial capacity, compared to 90% for δ-MnO₂.

As illustrated in Figure 8a, the discharge profiles reveal that the battery with a δ-MnO₂/Graphene cathode and a Zn-Al anode displays a discharge plateau around 1.63 V, while the same cathode paired with a TAl anode exhibited a plateau at 1.3 V. During charging, Zn²⁺ (produced by oxidation of Zn at the anode during discharge) and Al³⁺ ions co-deposit onto the zinc substrate, forming a Zn-Al alloy that causes a higher discharge voltage plateau and reduced polarization in the Zn-Al system. This alloy formation helps suppress the formation of an insulating Al₂O₃ layer and mitigates aluminum self-corrosion, both of which would otherwise contribute to increased overpotential, particularly due to the competing hydrogen evolution reaction [3].

Comparing the Zn-Al system to the battery with a TAl anode, the latter achieves a higher capacity (Figure 8a). In the Zn-Al system, zinc lowers the total amount of aluminum that is electrochemically active. While zinc forms a large part of the alloy, it does not contribute to the battery’s capacity. The lower aluminum proportion in the alloy results in fewer active aluminum atoms available for charge storage compared to the pure aluminum system, where 100% of the anode material consists of aluminum, leading to a higher total capacity. Due to the solubility limit of aluminum sulfate in water (a maximum of 0.5 M; exceeding this concentration results in precipitates), the initial concentration of Al³⁺ in the electrolyte is 1 M. However, the effective concentration of available Al³⁺ during cycling is lower, as a portion is consumed in alloy formation with Zn at the anode. Consequently, the remaining amount of Al³⁺ in the electrolyte is reduced, which may also contribute to the decrease in capacity. Moreover, certain side reactions, such as hydrogen evolution or self-corrosion of aluminum, may contribute to the total measured capacity, especially in early cycles.

Three-electrode CVs were conducted (Figures S6 and S8), confirming that the electrolyte remains stable within the operating voltage window of the TAl and Zn–Al batteries. However, the CV of the electrolyte from −1.0 to +0.6 V vs. Hg/Hg₂SO₄ in Figure S8 shows an anodic redox process emerging above ~+0.5 V that is attributable to the oxidation of Mn²⁺ to MnO_x/MnO₂ on the cathode during charge [70]. This feature is absent when the sweep is limited to +0.5 V. Using the potential–cell-voltage mapping in Figures S6 and S8, the onset corresponds to ~1.5 V in the TAl cell and ~1.9 V in the Zn–Al cell; thus, in the presence of Mn²⁺ (from MnSO₄), anodic MnO₂ electrodeposition is expected only when the positive electrode is driven beyond these voltages. Consistent with this, in Figure 8b, the TAl|δ-MnO₂ cell reaches ~1.6 V during charging, where MnO₂ deposition on the δ-MnO₂ electrode may occur, whereas the Zn–Al|δ-MnO₂ cell remains below ~1.9 V, making such deposition unlikely.

Although MnO₂ deposition is unlikely to occur in our Zn–Al system, Mn²⁺ may still contribute to charge storage alongside Al³⁺. Upon discharge, both cations may form the phase Al_xMn_1−xO₂, as proposed by He et al. [6], who observed a conversion/dissolution–deposition process involving Mn²⁺ and AlₓMnO₂. Notably, their system was charged beyond 1.5 V in an Al-anode configuration, which could allow MnO₂ deposition. Additionally, in our system, no evidence of new MnO₂-related morphology was observed on the cathode, and the layered morphology of the birnessite was retained after cycling. Therefore, although we do not claim MnO₂ deposition as an active mechanism in our system, we also cannot rule it out, as additional experiments would be required for confirmation.

Furthermore, CV data (Figure 7e,f) show two prominent redox peaks at ~1.6 V and 1.9 V for δ-MnO₂ and δ-MnO₂/Graphene coupled with a Zn-Al anode, closely matching voltages reported by Yan et al. [8] and Meng et al. [69]. for reversible Al³⁺ intercalation/deintercalation into AlₓMnO₂. Notably, the first CV cycle differs significantly from subsequent ones, confirming an initial irreversible conversion reaction. The close overlapping of the second and third cycles indicates a stabilized and reversible redox process. These features are consistently observed in both δ-MnO₂ and δ-MnO₂/Graphene electrodes. Taken together, our results are consistent with literature reports, suggesting that the dominant charge storage mechanism in our system involves an initial irreversible MnO₂ conversion, followed by reversible Al³⁺ intercalation/deintercalation into amorphous AlₓMn_(1−x)O₂. However, it should be noted that the detailed charge storage mechanism is not completely elucidated and requires additional merits.

As shown earlier in Figure 5 and Figure 7, the Zn–Al anode-based battery demonstrated superior cycling stability and higher coulombic efficiency compared to the TAl anode system, likely due to the suppression of side reactions such as hydrogen evolution. The introduction of additional metals to interact with aluminum is reported to alter its electrochemical behavior, potentially enabling underpotential deposition and promoting reversible metal deposition and dissolution at the anode, as suggested in previous studies [8,71].

In AAIBs, the co-deposition of Al and Zn involves the simultaneous reduction of Al³⁺ and Zn²⁺ ions onto a substrate, forming a Zn–Al alloy layer that enhances electrochemical performance. The deposition process begins with preferential Zn deposition due to its higher reduction potential relative to Al. The deposited Zn then facilitates Al nucleation by providing a conductive and chemically compatible substrate, lowering the energy barrier for Al deposition. Alloying between Zn and Al modifies the electrochemical environment at the electrode surface, mitigating passivation effects from aluminum oxide/hydroxide layers and suppressing the competing hydrogen evolution reaction [1,66]. Figure 9 shows a schematic representation of the suggested electrochemical reactions taking place in the δ-MnO₂/Graphene|Zn–Al cell.

Figure 8. (a) Charge/discharge profiles of the second cycle at 180 mA g⁻¹ for full cell batteries δ-MnO₂/Graphene cathodes using Zn-Al and TAl anode. (b) Potential of the working electrode versus time during charge/discharge for δ-MnO₂ cathodes with TAl and Zn–Al anodes. MnO₂ deposition on the anode may occur in the δ-MnO₂|TAl cell due to charging overpotentials reaching ~1.6 V, while it does not occur in the δ-MnO₂|Zn–Al system, where voltages remain below 1.9 V throughout cycling. (c) Comparing the electrochemical properties (capacity and voltage) of this study with those of recently reported AAIBs [1,6,8,9,10,11,12,22,23,72,73].

Figure 8. (a) Charge/discharge profiles of the second cycle at 180 mA g⁻¹ for full cell batteries δ-MnO₂/Graphene cathodes using Zn-Al and TAl anode. (b) Potential of the working electrode versus time during charge/discharge for δ-MnO₂ cathodes with TAl and Zn–Al anodes. MnO₂ deposition on the anode may occur in the δ-MnO₂|TAl cell due to charging overpotentials reaching ~1.6 V, while it does not occur in the δ-MnO₂|Zn–Al system, where voltages remain below 1.9 V throughout cycling. (c) Comparing the electrochemical properties (capacity and voltage) of this study with those of recently reported AAIBs [1,6,8,9,10,11,12,22,23,72,73].

Figure 9. Schematic representation of the electrochemical reactions occurring in the Zn–Al|0.5 M Al₂(SO₄)₃, 0.4 M MnSO₄|δ-MnO₂/Graphene system after the first cycle.

Figure 9. Schematic representation of the electrochemical reactions occurring in the Zn–Al|0.5 M Al₂(SO₄)₃, 0.4 M MnSO₄|δ-MnO₂/Graphene system after the first cycle.

Compared to reported AAIBs, the obtained specific capacity (746 mAh g⁻¹) ranks among the highest values achieved to date.

Table 2. The electrochemical characteristics of recently published AAIBs based on oxide cathode materials are contrasted with the findings of this study.

No.	Cathode	Anode	Electrolyte	Voltage (V)	Capacity (mAh g−1), Cycle Number	Ref.
1	AlxMnO2	Zn-Al	2 M Al(OTF)3	1.6	460, 80	[8]
2	V2O5@MXene	TAl	5 M Al(OTF)3	1	626, 100	[72]
3	MnO2	Zn-Al	3 M Al[TFSI]3	1.75	450, 400	[1]
4	AlxMnO2	a-Al	0.5 M Al2(SO4)3	1.8	<60, 200	[9]
5	AlxMn(1−x)O2	TAl	2 M Al(OTF)3, 0.5 M MnSO4	1.35	320, 65	[6]
6	AlxMnO2/C	MXene/ EAl97Ce3	2 M Al(OTF)3,0.2 M Mn(OTF)2	1.4	306, 500	[10]
7	MnAlxO2	Al	16.57 M AlCl3, 3.86 M MnSO4	1.9	285, 500	[23]
8	V6O13	Al	3 M Al(OTF)3	0.7	100, 1400	[73]
9	AlxMnO2	EAl82Cu18	2 M Al(OTF)3	1.5	400, 400	[12]
10	V-δ-MnO2	Al alloy	2 M Al(OTF)3	1.14	518, 400	[11]
11	Ov-MnO2	TAl	5 M Al(OTF)3	1.5	128, 200	[22]
12	AlxMn(1−x)O2/Graphene	Zn-Al	0.5 M Al2(SO4)3, 0.4 M MnSO4	1.63	746, 352	This work

and Figure 8c summarize the electrochemical properties of AAIBs batteries with oxide-based cathodes from recent literature, benchmarking them against the results of this study.

The Nyquist plots in Figure 10a,b, along with the fitting parameter values in

Table 3. Parameters of the equivalent circuit obtained from EIS data shown in Figure 10a,b, analyzed with EC-Lab software.

Cathode	R1 (Ω)	R2 (Ω)	R3 (Ω)	Q1 (μF.s1−n)	Q2 (μF.s1−n)
δ-MnO2 After 3rd cycle	4.12 ± 0.32	7.1 ± 0.4	3891 ± 100	87.7 ± 5.1	42.1 ± 3.4
δ-MnO2/Graphene After 3rd cycle	3.9 ± 0.21	4.4 ± 0.1	3055 ± 63	3.1 ± 0.4	69.5 ± 1.2
δ-MnO2 After 100th cycle	7.40 ± 0.05	47.2 ± 0.6	276 ± 3	26.7 ± 1.7	28.6 ± 0.4
δ-MnO2/Graphene After 100th cycle	6.70 ± 0.04	12.1 ± 0.1	136 ± 1	7.0 ± 0.1	141 ± 1

, reveal the electrochemical performance differences between δ-MnO₂ and its graphene composite (δ-MnO₂/Graphene) in a Zn-Al anode battery, demonstrating the enhanced electrochemical properties of the δ-MnO₂/Graphene composite. The same equivalent circuit used for the TAl system (Figure 5e) was applied to fit the Nyquist plots of the Zn-Al system after three cycles and after 100 cycles, with the difference after 100 cycles being the appearance of the Warburg element (W), which represents the impedance associated with solid-state ion diffusion into the bulk material. Similar to the impedance results observed in the battery with a TAl anode, the battery with a δ-MnO₂/Graphene cathode exhibits lower R values (Table 3). The δ-MnO₂/Graphene composite electrode demonstrates lower R values compared to δ-MnO₂ in a battery with a Zn-Al anode, even after 100 cycles, which reflects the composite stability over cycling. Additionally, the decrease in R₃ for both materials over 100 cycles, compared to R₃ after three cycles, suggests enhanced charge transfer, likely due to electrode surface activation during cycling.

The equation below can be used to calculate the bulk ionic diffusion coefficient (D_ion) [74,75]:

$$D_{ion}= {\displaystyle \frac{R^2{T}^2}{{2n}^4{A}^2{C}_{ion}^2{\sigma }^2}}$$

where R stands for the gas constant (8.314 J mol⁻¹K⁻¹), T for temperature (K), n for number of electrons transferred, A for electrode area (cm²), and C_ion for ion concentration (mol cm⁻³), and σ (Ω s^−1/2) refers to the Warburg coefficient extracted from the slope of the Z′ vs. ω^−1/2 plot (Figure 10c). Based on this equation, the lower σ observed for δ-MnO₂/Graphene indicates a higher D_ion, suggesting faster ion diffusion and better electrochemical performance. This can be attributed to the smaller MnO₂ nanoparticles (as determined by Scherrer analysis), which are anchored on a conductive graphene network, thereby reducing diffusion distances.

Comparison of the resistance values obtained from the Nyquist plots for batteries with Zn-Al and TAl anodes (Table 3 and Table 1, respectively) shows that the TAl system exhibits lower ohmic resistance (R₁) and charge transfer resistance (R₃). However, the Zn-Al alloy likely mitigates side reactions, such as hydrogen evolution and aluminum oxide formation, which are more pronounced in pure aluminum anodes. As a result, despite its higher charge transfer resistance, the Zn-Al system undergoes less voltage drop during discharge.

In contrast, while the TAl anode may initially offer more favorable charge transfer properties, it is more prone to degradation over time due to parasitic reactions. These effects reduce voltage stability and overall efficiency. Thus, although the Zn-Al system may have slower initial kinetics, it provides better energy retention, leading to a higher and more stable discharge plateau.

4. Conclusions

The incorporation of graphene into δ-MnO₂ significantly enhanced the electrochemical performance of AAIBs. XPS analysis confirmed the presence of oxygen vacancies (V_O) and Mn–O–C bonding in the δ-MnO₂/Graphene composite, both of which improve electronic conductivity and stabilize the MnO₂ during cycling. Notably, Mn–O–C bonding is known to mitigate Jahn–Teller distortions in Mn-based oxides, thereby preserving the integrity of the cathode material under repeated charge–discharge conditions, even though the δ-MnO₂ in the composite became XRD-amorphous during cycling.

In addition, defect-induced structural and electronic modifications, evidenced by the increased I_D/I_G ratio in the Raman spectra, further enhanced interfacial charge transport between MnO₂ and the graphene matrix. The EIS data also showed that the δ-MnO₂/Graphene electrode has reduced ohmic and charge transfer resistances. Furthermore, a lower Warburg coefficient (σ) is correlated with a higher ionic diffusion coefficient (D_ion), which indicates an improvement in ionic transport. Shorter diffusion pathways, due to smaller particle crystallite size, reduced particle agglomeration, and the open structure provided by graphene, further contributed to this improvement.

On the anode side, although TAl-based full cells initially exhibited higher capacities and lower charge transfer resistance, the introduction of Zn as a substrate for in situ Zn and Al co-deposition effectively extended cycle life and increased the discharge voltage to 1.63 V, compared to 1.3 V for the TAl system. The δ-MnO₂/Graphene|0.5 M Al₂(SO₄)₃, 0.4 M MnSO₄|Zn–Al system delivered an outstanding capacity of 746 mAh g⁻¹ over 352 cycles, while δ-MnO₂ alone retained a value of 642 mAh g⁻¹ for 107 cycles.

To the best of our knowledge, this study is the first to demonstrate a MnO₂–graphene cathode in AAIBs, highlighting the potential of graphene-based cathode composites for developing high-performance, cost-effective AAIBs, offering a promising pathway toward next-generation sustainable energy storage technologies.