Heterointerface Engineering‐Induced Oxygen Defects for the Manganese Dissolution Inhibition in Aqueous Zinc Ion Batteries

Manganese‐based material is a prospective cathode material for aqueous zinc ion batteries (ZIBs) by virtue of its high theoretical capacity, high operating voltage, and low price. However, the manganese dissolution during the electrochemical reaction causes its electrochemical cycling stability to be undesirable. In this work, heterointerface engineering‐induced oxygen defects are introduced into heterostructure MnO2 (δa‐MnO2) by in situ electrochemical activation to inhibit manganese dissolution for aqueous zinc ion batteries. Meanwhile, the heterointerface between the disordered amorphous and the crystalline MnO2 of δa‐MnO2 is decisive for the formation of oxygen defects. And the experimental results indicate that the manganese dissolution of δa‐MnO2 is considerably inhibited during the charge/discharge cycle. Theoretical analysis indicates that the oxygen defect regulates the electronic and band structure and the Mn‐O bonding state of the electrode material, thereby promoting electron transport kinetics as well as inhibiting Mn dissolution. Consequently, the capacity of δa‐MnO2 does not degrade after 100 cycles at a current density of 0.5 A g−1 and also 91% capacity retention after 500 cycles at 1 A g−1. This study provides a promising insight into the development of high‐performance manganese‐based cathode materials through a facile and low‐cost strategy.


Introduction
[3][4][5] Therefore, aqueous multivalent metal ion batteries (Zn, [6][7][8] Mg, [9] Al, [10] and Ca [11,12] ) have attracted much attention due to their electrolyte safety, simple preparation process, and higher theoretical capacity and energy density conferred by multielectron redox reactions.Among them, especially aqueous zinc ion batteries (ZIBs) have become a hot research topic in recent years due to their relatively low electrochemical reduction potential of zinc metal anodes (À0.763[15][16][17][18] Unfortunately, the capacity and cycle life of ZIBs assembled from various cathodes and zinc metal anodes have always struggled to meet the demands of practical applications.Hence, the development of cathode materials with high capacity and excellent cycling stability remains a major challenge. Currently, aqueous ZIBs cathode materials mainly include manganese-based, [19][20][21][22] vanadium-based, [23] Prussian blue analogs, [24] and organic materials. [25]In comparison, manganese-based cathode materials are considered to be a more promising cathode material for practical applications on account of their high operating voltage (1.3-1.4V), high theoretical capacity ( ∼ 308 mAh g À1 based on single electron transfer reaction), and highly tunable polycrystalline structure.However, the manganese dissolution and the irreversible structural degradation faced by manganese-based cathode materials in the electrochemical reaction make the electrochemical performance far from satisfactory. [26,27]To solve the above problems, a series of effective strategies have been gradually proposed, such as structural design, [28] complex construction, [29] intercalation tactics, [30] defect engineering, [31] and interfacial modification. [32]Given its role in modifying the surface chemistry and geometric morphology of oxides, oxygen defects are considered as an effective and feasible strategy for the development of high-performance ZIBs. [14]For example, Liang et al. prepared a potassium ion-stabilized and oxygen-deficient K 0.8 Mn 8 O 16 for an aqueous ZIBs cathode and found that the oxygen defects opened the MnO 6 polyhedral wall, thus facilitating the diffusion of the ions and suppressing manganese dissolution, which eventually resulted in a significant energy output of 398 W h kg À1 and an impressive durability over 1000 cycles. [21]Nevertheless, the introduction of oxygen defects in materials mostly required the assistance of high-temperature treatment or other complex processes, which would cause damage to the intrinsic structure, such as the loss of interlayer structural water.[35] However, the introduction of oxygen defects in certain materials using common electrochemical activation does not work as intended, and the regulation mechanism on the electrochemical performance by the oxygen defect requires further investigation.Recently, heterostructures have been proved to be very effective in the fields of catalysis and energy storage, and the interfacial effects between amorphous and crystalline domains perform a crucial role in the charge transfer/storage process. [36]Obviously, inducing the formation of oxygen defects by constructing heterointerface in electrochemical activation is extremely desirable.
Here, the heterostructure MnO 2 (δa-MnO 2 ) with abundant oxygen defects induced by amorphous-crystalline heterointerface in electrochemical activation was developed for aqueous zinc ion battery.This heterointerface was constructed by the inner amorphous MnO 2 core in concert with the outer crystalline MnO 2 nanosheet.In addition, interfacial effects at heterointerface were essential for inducing the formation of oxygen defects.Theoretical calculations revealed that oxygen defects could effectively modulate the electronic structure and Mn-O bonding state of the electrode material, resulting in significant enhancement of its structural stability.As a result, δa-MnO 2 not only exhibited enhanced electrochemical activity, but also the capacity degradation owing to manganese dissolution was remarkably suppressed.

Synthesis and Characterization
The fabrication of heterostructure oxygen-defect δa-MnO 2 was accomplished by a two-step process (Scheme 1).Firstly, amorphous-MnO 2 with a highly disordered structure was acquired by the redox reaction between a mixture of ethanol and ethylene glycol and potassium permanganate.The amorphous structure had substantial structural defects that could serve as reversible ion storage sites and might contribute to the mitigation of volumetric strain during cycling, thereby enhancing the structural stability of the electrode material. [37]Afterwards, δ-MnO 2 with layered structure was grown by in situ self-assembly on the above amorphous-MnO 2 surface using hydrothermal method, and then the amorphous-crystalline heterostructure was obtained.After that, the δa-MnO 2 was assembled with Zn anode and implemented electrochemical activation.Finally, the heterostructure oxygen-defect δa-MnO 2 was successfully synthesized.The construction of heterostructure would confer improved electrochemical activity as well as structural stability to the electrode material.
The morphology and structure of as-prepared samples were investigated using field emission scanning electron microscopy (SEM) and transmission electron microscopy (TEM).As a comparison, the SEM images of δ-MnO 2 showed a spherical morphology with a rough surface (Figure 1a; Figure S1a, Supporting Information).The energy barrier for homogeneous nucleation of δ-MnO 2 was high, and the spherical structure was obtained in order to minimize the nucleation energy of the system.Besides, its HRTEM images with an interplanar spacing of 0.408 nm corresponded well to the (002) crystal plane of layered manganese dioxide (Figure 1b; Figures S2 and S3, Supporting Information), which indicated that δ-MnO 2 had a distinct layered structure. [16]In addition, Figure S1b, Supporting Information displayed the irregular blocky morphology of amorphous-MnO 2 , and the magnification showed a smooth surface (Figure 1d).However, no significant lattice stripes were found in the HRTEM images of amorphous-MnO 2 (Figure 1e), suggesting that it had a highly disordered amorphous structure.Finally, the SEM images of δa-MnO 2 prepared by in situ selfassembly demonstrated a well-defined heterostructure (Figure S1c, Supporting Information), and the high magnification SEM images (Figure 1g) of its surface further revealed the abundant presence of homogeneous nanosheets.The morphological transformation of δ-MnO 2 after self-assembly on amorphous-MnO 2 was ascribed to the shift of its nucleation mode from homogeneous to heterogeneous nucleation, and the heterogeneous nucleation was more favorable to its nucleation. [38]dditionally, the heterointerface between crystalline and amorphous regions was clearly visible in the HRTEM image of δa-MnO 2 (Figure 1h; Figure S3, Supporting Information), which further proved the heterostructure oxygen-defect δa-MnO 2 were successfully obtained.The energy-dispersive X-ray spectroscopy (EDX) mappings of the above samples demonstrated that Mn and O elements exhibited a uniform distribution, which confirmed the presence of Mn and O elements in the as-prepared samples (Figure 1c,f,i).
The crystal structure of the as-prepared samples was further explored using XRD, as shown in Figure 2a.The sharp peaks at 12.1°, 24.5°, 36.7°, and 65.8°in the δ-MnO 2 diffraction pattern belonged to the (001), (002), (11-1), and (31-2) crystal planes of layered manganese dioxide (JCPDS 42-1317), respectively.On the contrary, the characteristic peaks of the (001) and (002) crystal planes in amorphous-MnO 2 disappeared, attributing to the formation of hydrated manganese oxide and the irregular arrangement of the MnO 6 octahedral skeleton. [37]Moreover, the peak intensities of the (11-1) and (31-2) crystal planes were drastically reduced, illustrating its highly disordered amorphous structural features.More importantly, the XRD pattern of δa-MnO 2 reappeared with the same characteristic peaks as that of δ-MnO 2 , further verifying that the nanosheets grown by self-assembly on the surface of amorphous-MnO 2 were layered manganese dioxide.The above XRD analysis results corroborated with SEM and TEM results.To reveal the chemical composition and elemental valence, the X-ray photoelectron spectroscopy (XPS) spectrum of the as-prepared samples is illustrated in Figure 2b-d.The overall XPS analysis (Figure S4, Supporting Information) demonstrated the primary peaks of Mn and O elements.The Mn 2p XPS spectrum (Figure 2b) displayed that the valence state of Mn element in δ-MnO 2 and amorphous-MnO 2 contained a small amount of trivalent in addition to the major tetravalent, but only tetravalent was available in δa-MnO 2 , implying that the heterostructure δa-MnO 2 had a higher average valence state. [39,40]The higher valence state of the heterostructure δa-MnO 2 was assigned to the strong interaction between the constructed heterointerface, [36,41] which promoted the stabilization of the valence state of Mn element and the formation of the higher valence state during the synthesis process.In addition, the splitting width between the two Mn 3s peaks of δa-MnO 2 was 4.81 eV (Figure 2c), which further evidenced that its Mn elemental valence state was tetravalent. [42][45] Obviously, the amorphous-MnO 2 promoted the stabilization of the valence state and avoided the influence of the samples from the external environment.In the O 1s XPS spectra (Figure 2d), the peaks located at 529.2, 530.7, and 532.2 eV were ascribed to lattice oxygen, oxygen defects, and adsorbed water, respectively. [46]Furthermore, there were more oxygen defects in both δa-MnO 2 and amorphous-MnO 2 than in δ-MnO 2 , which could be explained by the fact that they possessed more disordered amorphous regions.

Heterointerface-Induced Oxygen Defects
The zinc ion storage performance was investigated by employing the δa-MnO 2 cathode, Zn anode, 2 M ZnSO 4 with 0.1 M MnSO 4 aqueous electrolyte, and glass fiber separator in a coin cell configuration.After the first galvanostatic charge/discharge cycle at 0.2 A g À1 current density, the electrode material was sufficiently washed with deionized water to remove the electrolyte, and then the chemical environment of the elements in the electrode material was analyzed via XPS spectrum.Unexpectedly, a considerable increase in the oxygen defect content was observed in the high-resolution O 1s spectrum (Figure 3a) of the electrode material after the first charge-discharge cycle, which rose from 20.5% to 63.3% compared to the original δa-MnO 2 .To further corroborate the changes in the electrode material during the first charge-discharge cycle, the electron paramagnetic resonance (EPR) spectra (Figure 3b) was carried out to investigate the oxygen defect transition.Apparently, the signal of the electrode material at g = 2.00 after the first charge-discharge cycle was distinctly stronger than that of δa-MnO 2 before in situ electrochemical activation, demonstrating that the electrochemical activation boosted the transformation of oxygen defects, which was well consistent with the XPS results. [47]More importantly, the electron interaction between the oxygen defect and other atoms was the precondition for its ability to exist stably.The high-resolution O 1s spectra of the electrode material exhibited that the high content of oxygen defects survived even in the subsequent fifth charge-discharge cycle and did not disappear with the charge-discharge cycle (Figure 3c).The strong heterointerface interactions and the tuning of the electronic structure of the electrode material enabled the oxygen defects to be stable during the charge/ discharge cycles.Moreover, a weakening of the lattice oxygen peak after discharging from 1.8 to 0.8 V and an enhancement of the peak that possibly originated from absorbed H 2 O were found, which were assigned to the intercalation chemistry of Zn 2+ and H + . [48]As soon as it was charged to 1.8 V again, the above two peaks recovered reversibly, revealing the excellent reversibility of the charge/  discharge cycle of the electrode material.To examine the role of heterointerface in the introduction of oxygen defects, the O 1s spectra of δ-MnO 2 and amorphous-MnO 2 after the first charge/discharge cycle under the same conditions are shown in Figure 3d.After electrochemical activation, the oxygen defect content of δ-MnO 2 and amorphous-MnO 2 did not change much relative to their inactivated state (Figure 2d), which deviated greatly from the δa-MnO 2 electrode.As mentioned above, the heterointerface was essential in the construction of oxygen defects in the electrochemical activation of electrode materials.On the basis of the above results, a rational mechanism for heterointerface-induced oxygen defects was proposed.Generally, the outer layered nanosheet structure had a high crystallinity but inferior structural stability, which rendered it more prone to introduce crystal defects such as oxygen defects during the electrochemical activation. [33,49]However, the crystal defects with high energy hardly survived stably in the cycle, thereby failing to encourage the electrode material stability.As a result of interfacial interaction, the highly electrochemically active heterointerface effectively promoted the formation of oxygen defects and could stabilize them during electrochemical activation.According to the above analysis, it could be speculated that the introduction of high content oxygen defects would have a remarkable effect on the electrochemical performance of the electrode material (to be elaborated later).

Electrochemical Performance
The electrochemical performance of these samples was evaluated as cathode materials for aqueous ZIBs. Figure 4a shows the cyclic voltammetry (CV) curves of the δ-MnO 2 , δa-MnO 2 and amorphous-MnO 2 at a scan rate of 0.1 mV s À1 .Two groups of well-defined redox peaks were observed at around 1.53/1.25 V and 1.59/1.39V, which might stem from the intercalation of Zn 2+ and H + , [16] respectively.Meanwhile, the area of the closed region of the CV curve for δa-MnO 2 and δ-MnO 2 was notably larger than that of the amorphous-MnO 2 , reflecting their higher electrochemical reactivity, which was ascribed to the large layer spacing of the layered structure with more ion storage active sites.After in situ electrochemical activation, δa-MnO 2 exhibited dramatically improved reactivity compared to δ-MnO 2 , which was presumably dedicated to the introduction of abundant oxygen defects.The oxygen defects would greatly increase the donor density of δa-MnO 2 and would facilitate the electron transport and charge transfer during the redox reaction. [50]The discharge specific capacity of δa-MnO 2 was observed to be as high as 229 mAh g À1 in the galvanostatic charge/discharge (GCD) curves (Figure 4b) at 0.5 A g À1 current density, which was higher than that of δ-MnO 2 (208 mAh g À1 ) and amorphous-MnO 2 (177 mAh g À1 ).In addition, the GCD curves showed two distinct charging and discharging plateaus, corresponding to the two groups of redox peaks in the CV curves.In the GCD curves (Figure 4c) at different current densities, the discharge capacity of δa-MnO 2 reached 257 mAh g À1 at a current density of 0.1 A g À1 and retained 161 mAh g À1 even at 2 A g À1 .Although δ-MnO 2 (332 mAh g À1 ) performed higher capacity than δa-MnO 2 at a current density of 0.1 A g À1 ; however, the capacity decayed substantially to 95 mAh g À1 when the current density increased to 2 A g À1 , with a capacity retention of only 28% (Figure S5, Supporting Information).As shown in the rate performance of the as-prepared samples (Figure 4d), the capacity of δa-MnO 2 was slightly lower than that of δ-MnO 2 at a low current density, but the capacity of δ-MnO 2 declined dramatically as the current density rose rapidly, while δa-MnO 2 managed to maintain most of its capacity.
The long-cycle performance of all samples at different current densities is shown in Figure 4e,g.At a current density of 0.5 A g À1 , the specific capacity of δa-MnO 2 stayed at 175 mAh g À1 with almost no decay after 100 cycles, while δ-MnO 2 and amorphous-MnO 2 faded to 115 and 88 mAh g À1 , respectively.Even when the current density was raised to 1 A g À1 , δa-MnO 2 delivered 91% capacity retention compared to δ-MnO 2 (12%) and amorphous-MnO 2 (53%).Additionally, the electrochemical performance of δa-MnO 2 had a large advantage compared with that of the reported manganese-based cathode materials (Table S1, Supporting Information).As mentioned above, through the synergistic effect of the lattice modulation between the inner disordered amorphous structure and the outer layered structure as well as the in situ introduced oxygen defects, δa-MnO 2 not only achieved a high specific capacity but also a rather superior electrochemical stability.The electrochemical impedance spectroscopy (EIS) results demonstrated that δa-MnO 2 (19 Ω) had a smaller charge transfer resistance than δ-MnO 2 (132 Ω) and amorphous-MnO 2 (258 Ω) (Figure 4f), implying that the introduction of considerable oxygen defects substantially facilitated the diffusion process of Zn 2+ .
To explore the intrinsic mechanism of electrochemical performance enhancement, the changes of Mn element content in the electrolyte during the charge/discharge cycle at 0.5 A g À1 for all samples are shown in Figure 5a and Figure S6, Supporting Information.It could be seen that the Mn dissolution phenomenon was remarkably suppressed during the cycling of the δa-MnO 2 electrode; in contrast, the Mn content in the electrolyte of the δ-MnO 2 and amorphous-MnO 2 electrodes kept increasing, which was credited to the distinctive structural characteristics of δa-MnO 2 .It was also found that the amorphous-MnO 2 electrode appeared to separate the active material from the collector after 100 cycles, while the δa-MnO 2 electrode retained stability (Figure S7, Supporting Information).From the above result analysis, the manganese dissolution during the electrochemical reaction was effectively mitigated by the oxygen defects induced by amorphous-crystalline heterointerface, so that the capacity degradation was successfully suppressed.The improvement of electrochemical performances by manganese dissolution inhibition was further illustrated using the capacity vs. voltage (dQ/dV vs V) plots (Figure 5b).The redox peaks of δa-MnO 2 corresponding to the reversible intercalation and deintercalation reaction hardly changed in intensity even after 100 cycles, whereas the peak intensities of δ-MnO 2 and amorphous-MnO 2 decreased sharply after 100 cycles (Figure S8, Supporting Information), signifying that manganese dissolution inhibition effectively enhanced the cycling stability of the electrode materials.
In order to investigate the electrochemical kinetics of all samples, CV tests were first performed at scan rates of 0.1-1 mV s À1 .As shown in Figure S9a, Supporting Information, the peak currents of the two groups of redox peaks grew as the scan rate increased, but the shape of the CV curves remained constant.To gain insight into the ion storage mechanism of the electrode material, the ion storage behavior was further analyzed based on the relationship between the scan rate and the peak current [51] i ¼ aν b , ( where i and ν represent current density and scan rate, respectively.The b-value can be calculated by fitting the data using Equation (1) and varies between 0.5 and 1.When the b value tends to 0.5, it means that the ion storage behavior is diffusion-dominated; on the contrary, if the b value is closer to 1, it is a capacitance-dominated storage mechanism.According to Figure S9d, Supporting Information, the b-values of the four redox peaks were 0.63, 0.91, 0.88, and 0.69, respectively, illustrating that the charge storage behavior of the δa-MnO 2 electrode was controlled by a combination of diffusion and capacitive processes. [52]Both δ-MnO 2 (0.73, 0.95, 0.88, and 0.79) and amorphous-MnO 2 (0.61, 0.84, 0.82, and 0.64) electrodes were similar to the above fitting results for δa-MnO 2 (Figure S9b,c,e,f, Supporting Information).Also, the proportion of the capacitive contribution in charge storage using the following equation [51] I ν The k 1 ν and k 2 ν 1/2 correspond to the capacitive and diffusive contributions, respectively.The capacitance contributions of all samples were fitted and calculated (Figure 5c).The capacitance contributions of δa-MnO 2 were appreciably higher than those of δ-MnO 2 (75.3%, 78.5%, 83.3%, 85.6%, and 90.3%) and amorphous-MnO 2 (63.6%, 65.3%, 68.4%, 73.4%, and 84.5%) at different sweep rates.A high capacitive contribution evidenced that the electrochemical energy storage mechanism of the δa-MnO 2 electrode was dominated by the capacitive behavior.Apparently, the percentage of capacitance contribution of all samples increased with higher scan rate.The higher capacitance contribution of δa-MnO 2 could accelerate the electrochemical reaction kinetic of the electrode, which enabled a more excellent rate performance. [53]he galvanostatic intermittent titration technique (GITT) was employed to further investigate the electrochemical reaction kinetics of the electrode materials (Figure 5d).The ion diffusion coefficient (D) during the electrochemical reaction can be roughly estimated from the following equation [16] where D (cm 2 s À1 ), τ (s), m B (g), V M (cm 3 mol À1 ), M B (g mol À1 ), and A (cm 2 ) are the ion diffusion coefficient, relaxation time, mass of active material, molar volume, molar mass, and geometric area of the electrode, respectively.ΔE τ is the voltage difference caused by galvanostatic charging and discharging, and ΔE s is the voltage difference caused by pulsing.The relationship of the above variables is shown in Figure S10, Supporting Information.The ion diffusion coefficients (Figure 5e) of δa-MnO 2 were calculated to be in the scope of 10 À10 to 10 À11 cm 2 s À1 , much larger than those of δ-MnO 2 (10 À12 to 10 À13 cm 2 s À1 ) and amorphous-MnO 2 (10 À12 to 10 À13 cm 2 s À1 ).In accordance with the above electrochemical kinetic analysis results, the in situ construction of substantial oxygen defects could not only boost the charge transfer but also accelerate the substance transport.

Analysis of Energy Storage Mechanism
In-depth exploration of the charge/discharge mechanism of electrode materials has a profound significance for understanding the source of their excellent electrochemical performance.The ex situ XRD patterns of the δa-MnO 2 electrode at different charge and discharge states (0.8-1.8 V at 0.1 A g À1 ) for the second cycle are presented in Figure 6a.
With the continuous discharge of the electrode material, the XRD pattern gradually appeared three new diffraction peaks at 7.8°, 21.1°, and 58.3°, which were assigned to Zn 4 SO 4 (OH) 6 Á4H 2 O (ZSH, JCPDS:44-0673), and reached the maximum peak intensity at full discharge to 0.8 V. [54] The formation of ZSH resulted primarily from the dissociation of water molecules accelerated by the intercalation of H + during the discharge process, which led to an increase in the local electrolyte pH. [43]The SEM of the electrode surface during charging and discharging could also confirm the formation and disappearance of ZSH (Figure S11, Supporting Information).The surfaces of all samples fully discharged to 0.8 V were covered with apparent ZSH nanosheets, which was in agreement with the above XRD results.After recharged to 1.8 V, the ZSH nanosheets on the surface of δa-MnO 2 electrode completely disappeared, whereas a modest amount remained on the surface of amorphous-MnO 2 and δ-MnO 2 electrodes, which reconfirmed that the oxygen-rich defects conferred a superior electrochemical reaction reversibility to δa-MnO 2 .Besides, the other two diffraction peaks appearing at 32.6°and 34.7°were corresponding to tunnel-type ZnMn 2 O 4 (JCPDS:24-1133), which revealed the intercalation of Zn 2+ in the crystal structure of the electrode material. [48]When gradually charged from the fully discharged state, the intensity of the characteristic diffraction peaks attributed to ZSH and ZnMn 2 O 4 also slowly weakened until they disappeared, which reflected the deintercalation of Zn 2+ and H + .The reversible ion intercalation and deintercalation reaction demonstrated that δa-MnO 2 had outstanding reversibility of electrochemical reaction.Furthermore, the four diffraction peaks (12.1°, 24.5°, 36.7°, and 65.8°) belonging to the pristine electrode material could be maintained throughout the charge/discharge cycle, further manifesting the structural stability of δa-MnO 2 .To further verify the presence of Zn 2+ intercalation at the δa-MnO 2 electrode during the charging and discharging process, a mixture of 2 M ZnSO 4 and 0.1 M MnSO 4 and a solution containing only 0.1 M MnSO 4 were taken as the electrolyte for galvanostatic charge/discharge tests (Figure S12a,b, Supporting Information) at 0.5 A g À1 , respectively.When a mixture of 2 M ZnSO 4 and 0.1 M MnSO 4 was served as the electrolyte, the GCD curves of the δaMnO 2 electrode exhibited two distinct charge/discharge plateaus and a discharge capacity as high as 226 mAh g À1 , while only a steeply sloping slope and a discharge capacity of only 37 mAh g À1 were observed in 0.1 M MnSO 4 .From the result, it could be derived that the second charge/discharge plateau of the low potential originated from the intercalation and deintercalation of Zn 2+ . [42,55]ergy Environ.Mater.2024, 7, e12645 The valence changes of each element during the electrode charging and discharging were also examined in detail by the high-resolution XPS.With the discharge of the electrode from 1.8 to 0.8 V, a massive conversion of Mn (IV) to Mn (III) occurred (Figure 6b), which was responsible for the decrease of the valence due to the intercalation of Zn 2+ and H + . [56]After recharging to 1.8 V, the deintercalation of Zn 2+ and H + rendered Mn (III) oxidized to Mn (IV).In the high-resolution Mn 3s spectra (Figure S13, Supporting Information), the peak spin energy separation in the δa-MnO 2 electrode increased from 3.9 to 4.4 eV in the discharged state and decreased to 4.0 eV upon charging, which also illustrated that the valence state of the Mn element decreased upon discharge and increased upon charging. [44]As well, two Zn components, adsorbed Zn 2+ and intercalated Zn 2+ , were detected when the electrode was discharged to 0.8 V. [57] After fully charging, the intercalated Zn 2+ disappeared and the peak intensity of the Zn component was evidently reduced (Figure 6c), which was ascribed to the reversible deintercalation of Zn 2+ .Furthermore, Figure 6d demonstrates the effect of oxygen defects on the charge/discharge cycle of the electrode material.In contrast to the severe manganese dissolution during the charge/ discharge cycle in the pristine δa-MnO 2 electrode, the manganese dissolution was effectively suppressed by constructing abundant oxygen defects through in situ electrochemical activation, signifying that the oxygen defects imparted superior cycling stability to the electrode material by preventing manganese dissolution.
Analyzing the above results, the superior electrochemical performance of δa-MnO 2 could be ascribed to the following facts.At first, the combination of amorphous regions and ordered layered structure allowed the heterostructure structure of the electrode material to be susceptible to the formation of abundant and robust oxygen defects.60] Finally, the disordered amorphous cores with intrinsic defects could effectively release the structural stresses induced by ion intercalation and deintercalation, thus averting the structural degradation of the active materials. [61]5.DFT Calculations DFT calculations were carried out to gain insight into the mechanisms underlying the manganese dissolution inhibition and electrochemical performance enhancement by oxygen defects.Since the manganese dissolution of δa-MnO 2 during the charge/discharge cycles occurred majorly in the layered δ-MnO 2 in the surface layer, the optimized configurations of layered δ-MnO 2 with and without oxygen defects were constructed (Figure S14, Supporting Information).The total density of states (TDOS) and partial density of states (PDOS) of δ-MnO 2 and oxygen defect-MnO 2 are shown in Figure 7a,b.After the introduction of the oxygen defect, the DOS of MnO 2 shifted toward lower energy and the band gap decreased from 1.017 to 0.535 eV, which facilitated the leap of electrons from the valence band to the conduction band, leading to the enhancement of the intrinsically poor electronic conductivity of MnO 2 .The differential charge density (Figure 7c) revealed that the Mn atoms near the oxygen defect appeared to accumulate more charge, and the substantial charge redistribution would boost the electrochemical activity and interatomic bonding strength.Additionally, the bond length and bond angle of the Mn-O bond (Figure 7d) around the oxygen defects were reduced relative to those of MnO 2 without oxygen defects, which further suggested that the bond strength of the Mn-O bond was strengthened.In general, oxygen defects inhibited manganese dissolution by modifying the electronic structure and the bonding state of the Mn-O bonds, hence improving the cycling stability of the electrode material.

Conclusion
In summary, massive oxygen defects are induced by amorphouscrystalline heterointerface in electrochemical activation for aqueous Znion battery.The heterostructure consisting of the disordered amorphous MnO 2 core and the outer crystalline MnO 2 nanosheets plays a pivotal role in the construction of oxygen defects.Theoretical calculations demonstrate that the oxygen defect can not only accelerate the electron transfer efficiency by tuning the electronic structure and charge redistribution to enhance the electrochemical activity, but also tailor the bonding state of the Mn-O bond to mitigate the manganese dissolution to strengthen the structural stability.Based on these advantages, δa-MnO 2 delivers a capacity of up to 257 mAh g À1 at a current density of 0.1 A g À1 , and the capacity does not degrade after 100 cycles at 0.5 A g À1 .This work provides a simple and inexpensive strategy for the development of high-performance aqueous zinc ion battery cathode materials and may facilitate the application of structural refinement design to cathode materials development.

Experimental Section
Materials: Potassium permanganate (KMnO 4 ), ethanol, ethylene glycol, and hydrochloric acid (HCl, 38%) were purchased from Sinopharm Group Chemical Reagent Co., Ltd.All the chemicals were used as received without further purification.
Synthesis of the amorphous-MnO 2 : In a typical synthesis procedure, 1.1 mmol of potassium permanganate (KMnO 4 ) was added into 100 mL deionized water and stirred to fully dissolve to obtain solution A. Then 5 mL ethanol and 5 mL ethylene glycol were evenly mixed to obtain solution B. Subsequently, solution B was slowly dropped into solution A. After the dropwise addition, the above mixed solution was reacted in a water bath at 45 °C for 30 min, and the precipitate obtained from the reaction was centrifuged with deionized water for 3-5 times.Finally, the sample was freeze-dried sufficiently and noted as amorphous-MnO 2 .
Synthesis of the δa-MnO 2 : 0.2 mmol KMnO 4 and 1 mL hydrochloric acid were added into 60 mL deionized water to form a homogeneous solution.Then 100 mg amorphous-MnO 2 was added into the above mixed solution and reacted hydrothermally at 60 °C for 12 h.After the reaction was completed, the precipitate was centrifuged 3-5 times with deionized water and dried in air at 60 °C overnight to finally obtain δa-MnO 2 .
Synthesis of the δ-MnO 2 : 0.4 mmol KMnO 4 and 1 mL hydrochloric acid were added into 60 mL deionized water to form a homogeneous solution, then the mixed solution reacted hydrothermally at 60 °C for 12 h.After the reaction was completed, the precipitate was centrifuged 3-5 times with deionized water and dried in air at 60 °C overnight to finally obtain δ-MnO 2 .
Materials characterization: The microscopic morphology of the samples was observed by fieldemission scanning electron microscopy (FESEM, Hitachi S-4800) and energy-dispersive spectroscopy (EDS).The high-resolution TEM (HRTEM) images were obtained using transmission electron microscopy (TEM, JEM-2100F).The X-ray diffraction (XRD) patterns were measured on a Bruker D8 X-Ray diffractometer operated at 40 mA and 40 kV with Cu K α radiation at room temperature (λ = 1.5418Å).The elements and valence states on the carbon sphere are determined by X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB 250Xi), and the binding energies obtained in the XPS spectra were calibrated by the binding energy of C 1s at 284.5 eV.The electron paramagnetic resonance (EPR) tests were conducted in the X-band (9.45 GHz) with 5.00-G modulation amplitude using a Bruker EPR spectrometer (A300-10-12; Bruker).
Electrochemical measurements: δa-MnO 2 , ketjen black, and polyvinylidene fluoride were dispersed in N-methyl-2-pyrrolidone solvent in the ratio of 7:2:1.After thorough mixing, the above slurry was evenly coated on the stainless-steel mesh and dried at 60 °C for 12 h.The cells were assembled with δa-MnO 2 electrode as cathode, zinc foil (0.1 mm) as anode, Waterman filter paper as separator, and a mixture of 2 M ZnSO 4 and 0.1 M MnSO 4 as electrolyte in CR2032 coin cells, respectively.The cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) tests were conducted in a CHI660E electrochemical workstation (CHI 660E; CH Instruments, China).The galvanostatic charge/discharge (GCD), rate performance, long-cycle performance, and galvanostatic intermittent titration technique (GITT) measurements were performed on the NEWARE BTS-610 (Neware Co., Ltd., China) battery test system.
Calculation details: All density functional theory (DFT) calculations are performed using the Dmol3 package. [62]The electron exchange-correlation potential is described by the general gradient-corrected (GGA) with Perdew-Burke-Erzerhof (PBE) functional.The electrons of all atoms are described by DFT Semi-core Pseudopots (DSPP) basis set and a double numerical plus polarization (DNP).The atomic space cutoff radius is set to 4.4 Å.In addition, a thermal smearing (0.005 Ha) is exploited in order to accelerate the convergence speed during geometric optimization.The differential charge density is obtained according to the equation Δρ = ρ AB À ρ A À ρ B , where ρ AB , ρ A , and ρ B represent the charge density of oxygen defect-MnO 2 , MnO 2 without oxygen defects, and O atoms, respectively.

Figure 3 .
Figure 3. a) The high-resolution O 1s XPS spectra; b) EPR spectra of the δa-MnO 2 and after first cycle.c) The O 1s XPS spectra during the fifth charge/discharge cycle of δa-MnO 2 .d) The O 1s spectra of δ-MnO 2 and amorphous-MnO 2 electrodes after the first charge/discharge cycle.

Figure 4 .
Figure 4. a) The CV curves; b) the GCD curves at 0.5 A g À1 current density for all the prepared samples.c) The GCD curves of δa-MnO 2 at different current densities.d)The rate performance from 0.1 to 2 A g À1 for all the prepared samples.e) Cycling performance of for all the prepared samples at 0.5 A g À1 .f) EIS curves for all the prepared samples.g) Cycling performance of for all the prepared samples at 1 A g À1 current densities.

Figure 5 .
Figure 5. a) The changes in the content of Mn element in the electrolyte after different charge/discharge cycles of the samples at 0.5 A g À1 .b) dQ/dV plots of δa-MnO 2 and δ-MnO 2 .c) Contribution ratios of the capacitive capacities in the δ-MnO 2 , δa-MnO 2 , and amorphous-MnO 2 electrode.d, e) The GITT profiles and the corresponding ion diffusion coefficients of δ-MnO 2 , δa-MnO 2 , and amorphous-MnO 2 at charge/discharge states.

Figure 6 .
Figure 6.a) Charge/discharge curve of δa-MnO 2 electrode for the second cycle at 0.1 A g À1 and corresponding ex situ XRD patterns in different discharge/charge states.b, c) Ex situ high-resolution XPS spectra of (b) Mn 2p and (c) Zn 2p in different discharge/charge states.d) Schematic illustration of the effect of oxygen defects on the charge/discharge cycle.

Figure 7 .
Figure 7.The total DOS and PDOS of a) δ-MnO 2 , b) oxygen defect-MnO 2 .c) differential charge density of oxygen defect-MnO 2 .d) The changes of Mn-O bond length and bond angle before and after the introduction of oxygen defects.