The MnO₂-based nanostructures were prepared by hydrothermal method using KMnO₄ and MnSO₄ as starting materials. Pure MnO₂ was synthesized by dissolving a mixture of 3.2 g KMnO₄ and 5.07 g MnSO₄.H₂O in 40 mL deionized water. Ce-inserted MnO₂ was prepared using the same amounts of KMnO₄ and MnSO₄.H₂O in presence of 1 g Ce(NO₃)₃.6H₂O in 40 mL deionized water. The mixtures were stirred at 50°C–60 °C for 2 h. At the end of the stirring, the pink color characteristic of KMnO₄ disappeared, indicating the complete reduction of Mn(VII) and the oxidation of Mn(II) into Mn(IV), as shown in the equation below: 2   M n O 4 − a q + 3   M n 2 + a q + 2   H 2 O → 5   M n O 2 s + 4   H + a q

A dark precipitate of MnO₂ was formed during the reaction. The suspensions were transferred into separate Teflon jars, then enclosed inside stainless steel autoclaves. The tightly sealed autoclaves were transferred inside an oven and kept at 180°C for 24 h. After the hydrothermal treatment, the content of jars were naturally cooled down to room temperature. The solids were separated by centrifugation at 4000 rpm for 10 min at 20°C, and then washed several times by deionized water until a neutral pH was attained. The washed solids were dried in vacuum oven at 80°C overnight.

A slurry containing the active material, conductive material, binder, and solvent was prepared using a ball mill machine (Changsha Tianchuang Powder Technology Co., Ltd. XQM-0.4A), agate grinding Teflon jars and three agate balls (diameter 4 mm). In each jar, 0.75 g of the active material was mixed with 0.15 g of carbon (graphite) used as conductive material, and 0.1 g PVDF used as binder in 3.5 mL N-methyl-2-pyrrolidone (NMP). The agate grinding jar was placed inside the ball miller at 400 rpm for 1 h. The slurry obtained was casted on Ti foil by Doctor Blade and a thick film coater (MTI, MSK-AFA-I), followed by drying in the oven overnight. Then, the dried casted materials were cut into disks using a disk cutting machine (MTI, MSK-T10).

The prepared materials were tested as cathode material in half-cell using Zn metal as counter and reference electrode. The electrolyte used was 3 M ZnSO₄ aqueous solution. The coin cells (CR 2032) were sealed by a hydraulic crimping machine to finalize its assembly (MTI, MSK-110).

Cyclic voltammetry (CV) is a widely used technique in electrochemistry. CV is a powerful tool to study the electrochemical behavior of a system by systematic study of current-voltage measurements of a given electrochemical cell. The oxidation or reduction taking place at the electrode surface is related with the potential of the electrode. The more negative potential implies more strongly reducing electrode and on the other hand more positive potential implies more strongly oxidizing electrode. Figures 3A, B presents the consecutive cyclic voltammograms of MnO₂ and Ce-MnO₂ used as working electrodes in half-cell ARZIBs using Zn metal as reference and counter electrode and 3 M ZnSO₄ aqueous electrolyte. All peak positions identified for both half-cells are summarized in Table 1. Similar results were obtained in literature, which resembles the shape of CV curve in this study, as well as almost identical peak positions (Shi et al., 2020; Zang et al., 2020; Cai et al., 2021). In the return scan (reduction), two peaks appear at around 1.2 V and 1.3 V vs. Zn/Zn²⁺, which can be attributed to Zn²⁺ ions insertion into the MnO₂ host structure wherein Mn⁴⁺ is reduced to lower valences (Figure 3A) (Cai et al., 2021). The first oxidation scan of MnO₂ nanofibers cathode presents two peaks at approximately 1.6 V and 1.7 V vs. Zn/Zn²⁺, corresponding to the oxidation of Mn species back to Mn⁴⁺ and the de-insertion of Zn²⁺ ions. The only inconsistency lies in the cathodic region of the first CV scan, which showed a reduction peak positioned past 1.0 V, which also appeared for the Ce-MnO₂ cell. In addition, the intensity of the peaks from both oxidation and reduction have been shown to decrease after CV2. On a positive note, consistent overlapping of the peaks from CV3 until CV10 was also observed, which signifies good reproducibility for long-term cycling. The CV of Ce-MnO₂ in Figure 3B shows two peaks in the cathodic region at 1.2 and 1.3 V vs. Zn/Zn²⁺. Two consistent anodic peaks that are associated with the oxidation of Mn species back to Mn⁴⁺ were observed at 1.6 V (prominent) and 1.7 V (shoulder) vs. Zn/Zn²⁺. A peak at 1.1 V vs. Zn/Zn²⁺ was present in the first CV scan, which then shifted to around 1.2 V in the later scans. A slight increase in peak intensity was observed in the first two scans, which then remained consistent until CV 10. After CV 5, the peaks overlap which indicates favorable reproducibility upon further cycling. Generally, the number of peaks in the anodic and cathodic region are equivalent and not too far apart, indicating the reversible insertion/de-insertion of Zn²⁺ ions. Overall, both cells have shown good reproducibility and identical electrochemical peaks, indicating reversible insertion/de-insertion of Zn²⁺ ions. However, Ce-MnO₂ half-cell has shown better reproducibility with almost identical peak positions and intensities throughout the ten cycles.

For comparison, CV was repeated for MnO₂ and Ce-MnO₂ coin cells using 3 M Zn(CF₃SO₃)₂ as aqueous electrolyte (also known as Zn(OTf)₂). Cyclic voltammograms for each MnO₂ and Ce-MnO₂ half-cell against Zn/Zn²⁺ are presented in Figures 3C, D. The peak positions are summarized in Table 1 against that of ZnSO₄ electrolyte. The CV scan for MnO₂ nanofiber material (Figure 3C) displays two cathodic peaks positioned at 1.7 and 2.1 V. The second peak at 2.1 V only appears to be present from CV5 and beyond. As for the reduction, three peaks appear at 0.8, 1.3, and 1.9 V, with an additional shoulder peak appearing at 2.1 V. However, the first peak located 0.8 V gradually decreased in intensity with each CV scan and completely disappeared from CV5 and beyond. The peak intensities have increased from CV1, reaching similar intensities in CV5 and CV10, as evident in the overlap. In terms of reproducibility, CV scan for MnO₂ nanofiber material with ZnSO₄ electrolyte in Figure 3A has shown higher reproducibility than that of Zn(CF₃SO₃)₂ electrolyte in Figure 3C, which might be due to the secondary reactions of dissolution of Ti current collector in Zn(CF₃SO₃)₂ at high voltages. Nevertheless, more distinct peaks were achieved when Zn(CF₃SO₃)₂ electrolyte was used. The cyclic voltammogram of Ce-MnO₂ half-cell (Figure 3D) presents peaks at similar positions seen in Figure 3B (anodic: 1.6 and 1.7 V; cathodic: 1.1, 1.2, and 1.3 V). The peak intensities appear to decrease with longer cycling and stabilize between CV5 and CV10. Compared to Figure 3B, this CV scan is more consistent with the peak positions, which indicates good recyclability.

To evaluate the electrochemical performance of α-MnO₂ and Ce-inserted α-MnO₂ as cathode materials for aqueous ZIBs, half-cells were prepared and cycled in battery analyzers. The charge rate (current density) in the GCD cycling of MnO₂ and Ce-MnO₂ was fixed to 2500 mA/g. Figures 3E, F shows voltage profiles (changes of voltage with specific capacity) during consecutive GCD cycles of MnO₂ and Ce-MnO₂ cathode materials tested in half-cell ARZIBs using Zn metal as reference and counter electrodes and 3 M ZnSO₄ electrolyte. The first discharge and charge cycles of MnO₂ cathode materials present reduction and oxidation plateaus at 1300 mV and 1500 mV vs. Zn/Zn²⁺, respectively (Figure 3E). For the subsequent GCDs, the same plateaus exist with a decrease in the specific capacity from one cycle to another. A good accordance between the CV results and the GCD plots was perceived. The GCD cycles of Ce-MnO2 present two plateaus in charge and two plateaus in discharge. The GCD cycles of Ce-MnO₂ showed better stability (Figure 3F) than those of MnO₂.

Figures 3G, H present the changes of specific charge and discharge capacities and Coulombic efficiency (CE) in 500 cycles. The first specific charge and discharge capacities of 55 and 52 mA h/g for MnO₂, as shown in Figure 3G. The charge specific capacity undergoes a little increase and then stabilizes after 10 cycles, with CE reaching almost 100%. The specific charge capacity was retained at about 65 mA h/g for the rest of GCD cycles. Similar results were observed with Ce-MnO₂ cathode materials (Figure 3H). It is markedly noted that Ce-MnO₂ exhibited a more stable performance and much higher specific capacity values than MnO₂, with a specific capacity of approximately 115 mA h/g throughout the 500 cycles. The pre-insertion of Ce⁴⁺ ions into the tunnels of α-MnO₂ improves the specific capacity of the battery. The average specific charge capacities of Ce-MnO₂ after 10 GCD cycles at each of the following rates 100, 200, 500, 1000, 1500, 2000, 2500, 5000, 10000 were 297, 256, 222, 183, 153, 136, 116, 97, and 68 mA h/g, respectively. The Ce-MnO₂ retained an average specific charge capacity of 246 mA h/g at 200 mA/g indicating a robust chemical structure and excellent rate capability.

TEM analysis of the MnO₂ cathode supports the images produced from SEM, which revealed a nanofiber structure (Figure 5A). The crystal plane that was evident in all stages is (110), which has a d-spacing of 0.69 nm in α-MnO₂·0.15 H₂O. In Figure 5D, the structures formed during discharge in SEM imaging was also observed in TEM, with lengths of over 1 µm. Furthermore, a change from monocrystalline to polycrystalline phase was evident in the SAED pattern images before cycling (Figure 5C) and after discharge (Figure 5F). Besides (110) crystal plane, three other planes were observed including (020), (310), and (121), which are the main crystal planes of α-MnO₂·0.15 H₂O. After charge, a return to monocrystalline phase was observed, which revealed crystal plane (110). This indicates that the anode material is capable of reverting back to the original state before cycling, which is equivalent to good reversibility, as reflected in electrochemical findings.

The TEM imaging of Ce-MnO₂ cathode in post-mortem analysis showed the nanorod structures present in SEM analysis (Figure 6A). Similar to MnO₂, Ce-inserted

MnO₂ had d-spacing that matched well with that of α-MnO₂·0.15 H₂O, especially the characteristic (110) crystal plane. Before cycling, Ce-MnO₂ cathode was polycrystalline, as shown in the SAED pattern image in Figure 6C. After discharge, the phase changed to monocrystalline. The phase returned back to polycrystalline after charge, indicating good reversibility of the cathode.

Overall, post-mortem analysis has revealed that the cathode is capable of reverting back to the original structure, which is indicative of good reversibility. For both cathodes, the formation of microstructures were evident in SEM and TEM analysis in the discharged state. In particular, MnO₂ cathode showed a more significant change from nanofibers to stacked 3-D units. However, these structures were replaced with the original nanofibers and scattered microsheets, which were previously reported to be [Zn(OH₂)₃ZnSO₄⋅xH₂O]. In all states of the cathode, the (110) crystal plane was observed in HRTEM and SAED pattern images. Therefore, the original structure of MnO₂ was preserved in both pristine and Ce-inserted cathodes.