English

• Electrochemical energy storage technologies play an important role in updating the current energy infrastructure from clean and renewable power sources, such as solar and wind powers [1-3]. Among various energy storage candidates, rechargeable aqueous zinc-ion batteries (ZIBs) have received tremendous interest in scientific and technological communities owing to their advantages of high safety, low cost, eco-friendliness, and rich in natural resources [4-6]. Compared to the competing cathode materials of vandium-based compounds and prussian blue anologues, manganese oxides (e.g., MnO₂) stand out in terms of both high operating voltage (~1.3 V vs. Zn²⁺/Zn) and high specific capacity, making the Zn/MnO₂ battery chemistry particularly attractive [7-9].

  In recent years, a vast number of studies have developed various strategies, including nanostructuring, doping or pre-intercalation with metal cations or water, defect engineering, and electrolyte formulation, to modify the performance of MnO₂ [5, 10-12]. However, it is noted that the fundamental working mechanism of MnO₂ remains to be a controversial issue at present. Generally, four energy storage mechanisms have been proposed so far: (ⅰ) reversible (de)intercalation of Zn²⁺ [13-15], (ⅱ) conversion reaction of H⁺ [16], (ⅲ) co-(de)intercalation of Zn²⁺ and H⁺ [17-19], and (ⅳ) electrolytic deposition/dissolution of MnO₂ [20-22]. The former three ones are on the basis of one-electron Mn³⁺/Mn⁴⁺ reaction (theoretical capacity: 308 mAh/g), and the latter one is based on two-electron Mn²⁺/Mn⁴⁺ reaction (theoretical capacity: 616 mAh/g). Literature survey of various MnO₂ cathodes indicates an initial specific capacity of < 300 mAh/g, enabling the one-electron Mn³⁺/Mn⁴⁺ reaction mechanism more reasonable and popular [23-26]. Nevertheless, a few studies found that the fully-discharged cathode is merely composed of the pristine MnO₂ phase structure [20-22], which completely contradicts the formation of MnOOH or ZnMn₂O₄ products based on one-electron reaction. It is also worth to mention that zinc sulfate hydroxide (ZSH) phase will be formed on the cathode surface upon discharging [27, 28], which further complicates the discharged product identity. The ZSH is usually considered to be an undesirable by-product [16, 29, 30], the role of which, however, is still unclear during the charge/discharge processes. These discrepancies inspires us to present an in-depth understanding of the fundamental reaction mechanism and the associated ZSH issues.

  It is well-documented that Jahn-Teller distortion of Mn³⁺ will inevitably generate highly soluble Mn²⁺ species via a notable disproportionation reaction (i.e., 2Mn³⁺ → Mn²⁺ + Mn⁴⁺), thus resulting in structural instability regardless of the pristine crystal structure of α/β/δ/γ-MnO₂ [23, 31]. To mitigate the loss of active Mn from MnO₂, pre-addition of Mn²⁺ to the electrolyte with a proper concentration of 0.1–0.5 mol/L is regarded as the most promising way to date [16, 32]. Outstanding lifetime beyond thousands of cycles have been achieved at a high current rate ≥ 1 A/g in many previous studies [16, 26, 33-35]. Despite of these achievements, the cycling performance of MnO₂ at a low current rate of 100 mA/g is still a formidable challenge [5, 6]. Why the Mn²⁺ additive cannot solve the poor low-rate cycling problem? To answer this troublesome question, it is highly indispensable to unveil the role of Mn²⁺ and thus the capacity degradation mechanism.

  Bearing this motivation in mind, the present work aims to provide an in-depth insight into the fundamental mechanisms of redox reaction and cell degradation through a systematic electrochemical and structural investigation on α-MnO₂ crystalline nanowires. It is revealed that our Zn-MnO₂ cell experiences a two-electron reaction mechanism accompanying with a reversible dissolution/deposition of MnO₂ and precipitation/dissolution of flaky ZSH layer during discharge/charge processes. The ZSH plays a double-edged role: (ⅰ) Serving as a passivation layer to inhibit deep dissolution of MnO₂ upon discharging, and (ⅱ) promoting the electrochemical deposition kinetics of active MnO₂ upon charging. The addition of Mn²⁺ to the electrolyte could contribute supplementary capacity to maintain the cycling stability at high current densities, but a higher Mn²⁺ concentration (e.g., 0.5 mol/L) will expedite the cell failure by corroding the metallic zinc anodes. Moreover, the corrosion of zinc anode and the accumulation of less reversible ZnMn₂O₄ phases on the cathode collectively result in the cell degradation.

  MnO₂ materials were prepared by a simple hydrothermal method. Fig. 1 shows the physiochemical characterizations of the as-prepared sample. The sharp diffraction peaks in XRD pattern (Fig. 1a) can be readily assigned to the high-pure tetragonal α-MnO₂ phase with a typical [2 × 2] tunnel structure [36]. The structural information is further confirmed by Raman spectrum (Fig. S1 in Supporting information), in which the distinctive Raman bands at 583 and 641 cm⁻¹ belong to the stretching symmetric Mn-O vibrations of [MnO₆] basic slabs while those at 184 and 389 cm⁻¹ correspond to the deformation modes of Mn-O-Mn chains [18, 37]. The SEM (Fig. 1b) observation indicates that the MnO₂ material is typical of one-dimensional nanowire morphology with smooth surface. TEM imaging (Fig. 1c) validates the straight nanowire feature having an average diameter of ca. 100 nm. The selected area electronic diffraction (SAED) of a single nanowire displays a set of well-defined pattern spots, indicating high-quality single crystalline characteristics of the α-MnO₂. The HRTEM (Fig. 1d) exhibits clear lattice fringes with a distance of 0.49 nm, agreeing well with the interplanar spacing of (200) planes. The MnO₂ nanowires possess a BET specific surface area of 17.9 m²/g (Fig. S2 in Supporting information). The surface chemistry of the sample was examined by XPS technique. As shown in Fig. S3a (Supporting information), the survey spectrum reveals the presence of Mn, O and K elements. The K presence can be attributed to the incorporation of K⁺ cations into the [2 × 2] tunnels for stabilizing the α-MnO₂ structure. The high resolution Mn 2p spectrum (Fig. S3b in Supporting information) displays Mn 2p_1/2 and Mn 2p_3/2 peaks at 654.7 and 642.9 eV, respectively, corresponding to a typical spin energy gap of 11.8 eV for MnO₂ [25]. The core-level O 1s spectrum (Fig. S3c in Supporting information) can be fitted by two species of tetravalent Mn-O-Mn component at 530 eV and hydrated trivalent Mn-OH component at 531.4 eV. According to the envelope proportion of Mn-O-Mn and Mn-OH components [38, 39], the average Mn valence of MnO₂ calculated to be +3.81, which is in good accordance with the value determined by the Mn 3s spectrum (Fig. S3d in Supporting information).

  Figure 1

  

  Figure 1.  (a) XRD pattern, (b) SEM, (c) TEM, SAED (inset) and (d) HRTEM images of MnO₂ sample.

  DownLoad: Full-Size Img PowerPoint

  The as-prepared α-MnO₂ materials were employed as the cathodes of ZIBs to evaluate their electrochemical properties in three different aqueous electrolytes of 2 mol/L ZnSO₄ (2 M ZS), 2 mol/L ZnSO₄ + 0.1 mol/L MnSO₄ (2 M ZS + 0.1 M MS) and 2 mol/L ZnSO₄ + 0.5 mol/L MnSO₄ (2 M ZS + 0.5 M MS). Fig. 2a, Figs. S4 and S5 (Supporting information) show the initial six CV curves. It is noted that the α-MnO₂ cathodes share an identical CV shape in the three electrolytes, indicating the same reaction chemistry. The electrochemical behavior is further confirmed by the galvanostatic charge/discharge profiles (Fig. 2b), in which the stepwise voltage plateaus are in close association with the insertion of H⁺ and/or Zn²⁺. However, there is no consensus on the detailed working mechanism of MnO₂ at present. A number of previous studies evidenced that the discharge process is a one-electron solid/solid reaction (Mn⁴⁺(s) ↔ Mn³⁺(s), e.g., MnOOH and ZnMn₂O₄ as the products) [16, 17]. Some recent research substantiated a purely two-electron solid-liquid reaction mechanism (Mn⁴⁺(s) ↔ Mn²⁺(l)) [20, 22]. Hence, it is highly indispensable to unveil the true mechanism, which will be specified in the following section.

  Figure 2

  

  Figure 2.  (a) CV curves, (b) galvanostatic charge/discharge profiles of MnO₂ cathodes in 2 mol/L ZnSO₄ + 0.1 mol/L MnSO₄ electrolyte. Cycling performance of cells at a current rate of (c) 0.1 A/g and (e) 1 A/g (200 µL electrolyte). (d) Effect of electrolyte amount on the specific capacity.

  DownLoad: Full-Size Img PowerPoint

  Fig. 2c exhibits the cycling performance of the ZIB cells at 0.1 A/g in different electrolytes. The specific capacity of the cell shows a continuous decrease in pure ZnSO₄ electrolyte, which can be ascribed to the structural instability caused by the John-Teller effect of Mn³⁺ that produces soluble Mn²⁺ via disproportionation reaction [21]. The addition of Mn²⁺ in the electrolyte is generally considered to be a feasible approach to stabilize the MnO₂ structure by suppressing the disproportionation reaction of Mn³⁺ [16, 40, 41]. However, in the electrolytes containing different MnSO₄ concentrations, it is interesting to notice that both cells show a sharp increase of the capacity in the initial 30 cycles and then a fast capacity degradation. In particular, the cell delivers an incredible specific capacity of 1240 mAh/g in 2 M ZS + 0.5 M MS, which is double of the theoretical value even based on two-electron reaction (616 mAh/g). This suggests that the capacity calculation based on the initial MnO₂ mass is absolutely inappropriate in electrolytes containing MnSO₄ additive, making the performance comparison unfair among different studies. The capacity gain is related to the electro-oxidation of Mn²⁺ into new active MnO₂ species, as confirmed by the short charging voltage plateau at > 1.8 V (Fig. 2b) and the increasing capacity for the carbon black electrode (Fig. S6 in Supporting information). The phenomenon indicates that the Mn²⁺ amount in the electrolyte makes a critical contribution to the total capacity. In addition to the MnSO₄ concentration, it is believed that the electrolyte volume is also an important parameter for the capacity delivery. As evidenced in Fig. 2d and Fig. S7 (Supporting information), the capacity summit increases with the increasing usage of the electrolyte volume. Therefore, for the Zn-MnO₂ battery system, it is necessary to point out the Mn²⁺ concentration, the electrolyte volume, and the MnO₂ loading in one cell to enable a meaningful electrochemical evaluation of the materials and a fair comparison with the documented literature [42].

  The galvanostatic charge/discharge profiles of the cell at different current densities in 2 M ZS + 0.1 M MS are displayed in Fig. S8 (Supporting information). At a high current rate of 2 A/g, the cell could deliver a higher capacity of 138 mAh/g in the electrolytes containing MnSO₄ than that in ZS electrolyte. In addition, the two-stage discharge plateau is clearly visible at lower current rates (< 0.5 A/g) and the contribution ration of the upper discharge plateau increases from 55.8% to 80.2% as the current rate rises from 0.1 A/g to 2 A/g (Fig. S9 in Supporting information), indicating that the lower discharge plateau is a kinetically-limited reaction. The capacity contribution from the two plateaus during the cycling test is further quantified in Fig. S10 (Supporting information). Notably, the contribution from the upper discharge plateau is increasing from 50% to 73% after 150 cycles, suggesting that the upper discharge process is more reversible than the lower counterpart. Since the less reversible lower discharge plateau could be suppressed at high current rates, it is expected that the cycling performance can be greatly improved at a high current rate of 1.0 A/g. As shown in Fig. 2e, the cell manifests excellent cycling performance in 2 M ZS + 0.1 M MS over 1100 cycles. That's the reason why many previous-reported studies could obtain outstanding cycling performance at high current densities. Herein, it should be noted that the cycling performance in 2 M ZS + 0.5 M MS is much poorer than that in 2 M ZS + 0.1 M MS. We ascribe it to the metallic Zn anode problem, which will be clarified in the subsequent degradation discussion.

  Ex situ XRD, SEM and TEM techniques were carried out to probe the intrinsic mechanism of MnO₂ by investigating the structural and morphological evolution during the first two cycles. When the cell is fully discharged to 1.0 V (point B in Fig. 3a), the peak intensity of α–MnO₂ weakens significantly, and the new set of diffraction peaks can be assigned to hydrated zinc sulfate hydroxide (ZSH, Zn(OH)₂SO₄·5H₂O) phase (Fig. 3b) [21, 32]. A thick layer of ZSH flakes is precipitated on the cathode surface (Fig. 3d and Fig. S11 in Supporting information), compared to the pristine nanowire morphology (Fig. 3c). The formation of ZSH phase is further confirmed by the homogeneous distribution of Zn, S and O elements throughout the flake (Fig. S12 in Supporting information). It should be mentioned that the complex diffraction peaks of ZSH phase are almost overlapping with that of MnOOH and ZnMn₂O₄, which usually conclude misleading information. For example, some weak diffraction peaks are usually identified as MnOOH (e.g., the peak at ca. 21°) or ZnMn₂O₄ (e.g., the peaks at ca. 32°−36°) in previous studies [13, 17]. TEM imaging is then conducted to clarify this discrepancy, and the result reveals that the fully-discharged sample is merely composed of single-crystalline α–MnO₂ nanowires (Figs. 3e and f). However, there are many dissolved areas in the nanowire body from the dark-field TEM image (Fig. S13 in Supporting information), indicating the discharge process is a dissolution process (i.e., MnO₂ → Mn²⁺). To further validate it, the ZSH layer was washed by diluted acetate acid. It is observed that only α–MnO₂ nanowires are present in the washed cathode (Fig. S14 in Supporting information).

  Figure 3

  

  Figure 3.  (a) The initial two charge/discharge profiles, (b) ex-situ XRD patterns of the MnO₂ cathodes collected from different states. SEM, TEM, SAED (inset), HRTEM, and EDX elemental mapping images of the cathodes at (c) pristine A state, (d–f) fully-discharged B state, (g–i) fully-charged C state, and (j–l) partially-discharged D state.

  DownLoad: Full-Size Img PowerPoint

  After fully charged (point C), the cathode recovers the α–MnO₂ phase yet with much weaker diffraction intensity (Fig. 3b). The ZSH flakes are dissolved from the cathode surface and the nanowires are restored as observed from the SEM image (Fig. 3g). Interestingly, the nanowire surface is covered by a layer of nanoporous nanosheets, the phase of which cannot be detected by the XRD technique. The unique core-sheath nanowire/nanosheet architecture is clearly visible by TEM imaging (Fig. 3h). The corresponding SAED analysis shows two sets of diffraction patterns, in which the disperse rings belong to the low-crystalline birnessite-type δ-MnO₂. The lattice spacing of 0.69 nm in the HRTEM (Fig. 3i) corresponds to the (002) planes of birnessite-type δ-MnO₂, revealing that the charging process is an electro-oxidation of Mn²⁺ into birnessite-type δ-MnO₂ accompanying with the ZSH dissolution. The discharging process in the second and following cycles involves a two-stage reaction. As displayed in Figs. 3j and k, the cathode maintains the hybrid morphology of nanowire/nanosheet after the upper discharge plateau (point D). In addition, a small amount of ZSH nanoflakes start to precipitate on the surface. The EDX elemental analysis of the nanosheets illustrates a higher Zn content of 13.63% (area 2, Fig. 3l) compared to the naked nanowire (area 1, 0.54%), indicating that the birnessite-type δ-MnO₂ is more energetically favorable for the insertion of Zn²⁺ ions than the α–MnO₂. A deeper discharging to 1.0 V (point E) results in the growth of microsized ZSH flakes and the dissolution of the outer MnO₂ nanosheets (Fig. S15 in Supporting information). And the subsequent re-charging process achieves the re-generation of nanowire/nanosheet morphology and the removal of ZSH flakes (Fig. S16 in Supporting information), indicating a highly reversible reaction.

  The above mechanistic discussion herein substantiates a two-electron solid-liquid MnO₂/Mn²⁺ reaction pathway accompanying with a reversible precipitation/dissolution of a ZSH layer. However, it is noted that the initial specific capacity of MnO₂ is much lower than the theoretical value based on the two-electron reaction, indicating that part of MnO₂ (~45%) participates in the discharge process. The reason can be attributed to the formation of the thick ZSH inert layer covered on the electrode surface, which passivates the deep reduction dissolution of MnO₂. This is proved by the EIS (Fig. 4a), from which the charge-transfer resistance (R_ct) of the fully-discharged cell is measured to be as high as 2610 Ω. When the ZSH layer is removed, the R_ct of the re-assembled cell (electrolyte: 2 M ZS + 0.1 M MS) is remarkably reduced to 405 Ω. More importantly, the re-assembled cell could deliver a discharge capacity of 147 mAh/g based on the initial MnO₂ mass (Fig. 4b), corresponding to 267 mAh/g based on the remnant MnO₂ mass. The result validates the passivation role of the ZSH layer on impeding further discharge of MnO₂. In addition, if the re-assembled cell is charged at first, the charge plateau in the voltage range of 1.5–1.6 V disappears, and a higher potential (> 1.8 V) is required to drive the oxidation of Mn²⁺ in the absence of ZSH, suggesting that the ZSH layer is of critical benefit for the electro-deposition of MnO₂. To further discuss the MnO₂/Mn²⁺ reaction, the fully-discharged cathodes were taken out to re-assemble new cells using fresh electrolytes with and without MnSO₄ additive. As shown in Fig. 4c, the cell cannot operate in the electrolyte without Mn²⁺. By contrast, normal charge/discharge profiles are obtained in the electrolyte containing Mn²⁺, confirming the dissolution/deposition mechanism. Moreover, the direct oxidation of Mn²⁺ in the absence of ZSH is also proved by the cell using carbon black (CB) as the cathode in the electrolyte containing MnSO₄ (Fig. S6). It is worth pointing out that the oxidation of Mn²⁺ is a kinetically sluggish process. As shown in Fig. S6b, the CB-cell exhibits increasing specific capacity at a low current rate of 0.1 mA/cm², but negligible capacity at a high current rate of 1 mA/cm², agreeing well with the earlier result in Fig. 2c.

  Figure 4

  

  Figure 4.  (a) EIS spectra of the fully-discharged cathodes before and after acid washing. (b) Charge/discharge curves of the cells re-assembled with the acid-washed cathodes at 0.05 A/g. (c) Charge/discharge curves of the cells re-assembled with the discharged cathodes without acid washing. (d) Schematic illustration of working mechanism of α-MnO₂ cathodes.

  DownLoad: Full-Size Img PowerPoint

  Based on the in-depth structural and electrochemical analysis, Fig. 4d illustrates the working mechanism of Zn/MnO₂ battery. The first discharge/charge stages Ⅰ/Ⅱ in Fig. 4d indicate a partial dissolution process of α-MnO₂, which can be written as [20]:

  (1)

  The ZSH phase is usually regarded as undesirable by-products [43]. Nevertheless, the formation of ZSH phase is double-edged, which not only functions as a passivation layer to impede deep dissolution of MnO₂ upon discharging, but also promotes the electro-deposition kinetics of birnessite-type MnO₂ upon charging. Additionally, the MnSO₄ additive in the electrolyte can be electrochemically oxidized into new active species at a higher potential of > 1.8 V (stage Ⅲ, i.e., Mn²⁺ + 2H₂O→ MnO₂ + 4H⁺), thereby making an extra contribution to the total capacity.

  In the following discharge/charge cycles, the as-deposited birnessite-type δ-MnO₂ could provide large layered channels for insertion of Zn²⁺/H⁺ with a smaller overpotental (stage Ⅳ):

  (2)

  The subsequent lower discharge plateau (stage Ⅴ) is similar to the Eq. 1, which is a typical dissolution of (Zn/H)xMnO₂ along with a simultaneous precipitation of ZSH flakes.

  As discussed in Fig. 2c, regardless of the electrolyte containing MnSO₄ or not, the ZIB cells can retain no more than 20 mAh/g after 200 cycles at a low current density of 0.1 A/g. The previous studies generally ascribe the poor cycling performance of MnO₂ to its structural instability caused by Jehn-teller effect [16, 21]. To this end, a great number of design strategies, such as pre-intercalation/doping of metal ions (e.g., K⁺, Zn²⁺, Ca²⁺, Co²⁺) and surface coating, have been proposed to stabilize the MnO₂ crystal framework [26, 33, 37, 44-46]. Although improved cycle life was obtained, these achievements are usually based on the tests at high current densities (≥1 A/g). Scarcely do these studies present the cycling performance measured at < 0.3 A/g beyond 100 cycles. In fact, from the perspective of the dissolution/deposition mechanism, it is meaningless to stabilize the MnO₂ matrix as the MnO₂ will inevitably suffer from structure reconstruction at each cycle. Hence, it is necessary to unveil the intrinsic degradation nature of the Zn/MnO₂ battery system.

  Fig. 5a shows the XRD pattern of the cycled MnO₂ cathodes. The peak intensity of α-MnO₂ is greatly decreased primarily due to the partial dissolution of pristine α-MnO₂ upon discharging and the surface deposition of amorphous birnessite-type MnO₂ upon charging. The SEM image (Fig. 5b and Fig. S17 in Supporting information) confirms the formation of an amorphous layer and some underneath α-MnO₂ nanowires can still be observed. In addition to the α-MnO₂, there are some strong peaks that can be assigned to the ZnMn₂O₄ phase. Compared to MnO₂, ZnMn₂O₄ phase has been reported to be less electrochemically acive for ZIBs [47, 48]. Guo et al. considered the main reason of capacity degradation as the formation of ZnMn₂O₄ [20]. However, in our work, when the cycled MnO₂ is re-assembled with a fresh Zn anode, the new cell could deliver a high and stable specific capacity of ~300 mAh/g at 0.1 A/g (Fig. 5c). Consequently, we believe that the capacity degradation is also in close relation to the Zn anode.

  Figure 5

  

  Figure 5.  (a) XRD pattern and (b) SEM image of the MnO₂ cathodes after the cycling test at 0.1 A/g. (c) Cycling performance of the cycled MnO₂ cathode re-assembled with a fresh Zn anode, and the inset shows the corresponding charge/discharge curves. (d) XRD of the Zn anodes in different electrolytes. (e–g) SEM images of the Zn anodes after soaking in different electrolytes for 10 days. (h) Long-term cycling stability of the Zn/Zn symmetric cells in different electrolytes.

  DownLoad: Full-Size Img PowerPoint

  XRD analysis of the cycled Zn anodes reveals the formation of the ZSH phase (Fig. 5d). Notably, the peak intensity of the ZSH increases with the increasing concentration of MnSO₄, suggesting that the MnSO₄ additive exacerbates the corrosion problem of Zn anode. Fig. S18 (Supporting information) shows the optical photos of the cycled Zn anodes. It is observed that the Zn plate suffers from more severe corrosion issue in the electrolyte of 2 M ZS + 0.5 M MS than that in 2 M ZS + 0.1 M MS. To further probe the chemical stability of Zn anodes, pure Zn plates were soaked in three different electrolytes for 10 days. The significant color change of the three Zn plates indicates a serious corrosion reaction (Fig. S19 in Supporting information). XRD identifies the formation of ZSH by-products, the peak intensity of which exhibits a similar increasing trend to the cycled anodes. The morphology of the Zn surface is compared by SEM imaging. Microsized ZSH particles are sparsely covered on the Zn surface after soaking in pure 2 M ZS electrolyte (Fig. 5e). The addition of MnSO₄ obviously aggravates the precipitation of ZSH particles (Figs. 5f and g). Remarkably, a high concentration of 0.5 mol/L MnSO₄ incurs a dense and thick ZSH flakes (Fig. 5g), validating the adverse role of the MnSO₄ additive on the stability of metallic Zn. Zn/Zn symmetric cells were assembled using the three different electrolytes to evaluate the long-term galvanostatic cycling stability at 1 mA/cm² (Fig. 5h). The cells show similar charge/discharge voltage profiles (Fig. S20 in Supporting information). The cells using both 2 M ZS and 2 M ZS + 0.1 M MS could maintain excellent cycling durability for more than 1050 h. In sharp contrast, the cell based on 2 M ZS + 0.5 M MS experiences inferior cycle life with a sudden reduction of the polarization voltage after 340 h. The cell decay is attributed to the dendrite-induced short circuit and accumulation of detrimental by-products [49-52]. As such, the worse cycling performance of the Zn/MnO₂ full cell in 2 M ZS + 0.5 M MS at the high current rate of 1 A/g (Fig. 2e) can be well rationalized by the poor electrochemical stability of Zn anode.

  In summary, the fundamental mechanisms of redox reaction and degradation based on aqueous Zn/α-MnO₂ battery system is unveiled by an in-depth electrochemical and structural analysis. The working mechanism involves a two-electron solid-liquid redox reaction of MnO₂/Mn²⁺ from part of MnO₂ along with a concomitant precipitation/dissolution of a flaky ZSH layer during discharge/charge processes. The ZSH by-product is double-edged, forming a passivation layer to inhibit deep dissolution of MnO₂ upon discharging, but kinetically promoting the reversible deposition of MnO₂ upon charging. The capacity degradation is primarily ascribed to the anode failure induced by metallic zinc corrosion and the accumulation of irreversible ZnMn₂O₄ phases on the cathode. The employment of electrolytic MnSO₄ additive can contribute additional capacity to the total capacity by electro-oxidation of Mn²⁺, but cannot solve the cycling problem at a low current density (e.g., 0.1 A/g). A high MnSO₄ concentration (e.g., 0.5 mol/L) greatly exacerbates the corrosion of the metallic zinc anode, thereby downgrading the cycling stability of full cells. Moreover, to enable a fair and meaningful performance comparison, the critical cell parameters of Mn²⁺ concentration, electrolyte amount, and the initial MnO₂ loading, should be provided as a protocol for evaluating Mn-based materials. It is believed that the fundamental study will shed light on the development of novel strategies toward a practically viable Zn/MnO₂ battery technology.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  The authors would like to acknowledge the research fund of National Natural Science Foundation of China (No. 51821091), and Fundamental Research Funds for the Central Universities (Nos. D5000210894 and 3102019JC005). The authors also thank the support from the Analytical & Testing Center of Northwestern Polytechnical University for TEM analysis.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.05.039.

  
  
• 1. [1]

     B. Dunn, H. Kamath, J.M. Tarascon, Science 334 (2011) 928–935. doi: 10.1126/science.1212741

  2. [2]

     Z. Yang, J. Zhang, M. Kintner-Meyer, et al., Chem. Rev. 111 (2011) 3577–3613. doi: 10.1021/cr100290v

  3. [3]

     S. Chu, Y. Cui, N. Liu, Nat. Mater. 16 (2016) 16–22.

  4. [4]

     J. Ming, J. Guo, C. Xia, et al., Mater. Sci. Eng. R 135 (2019) 58–84. doi: 10.1016/j.mser.2018.10.002

  5. [5]

     H. Liu, J.G. Wang, Z. You, et al., Mater. Today 42 (2021) 73–98. doi: 10.1016/j.mattod.2020.08.021

  6. [6]

     L.E. Blanc, D. Kundu, L.F. Nazar, Joule 4 (2020) 771–799. doi: 10.1016/j.joule.2020.03.002

  7. [7]

     M. Song, H. Tan, D. Chao, et al., Adv. Funct. Mater. 28 (2018) 1802564. doi: 10.1002/adfm.201802564

  8. [8]

     A. Konarov, N. Voronina, J.H. Jo, et al., ACS Energy Lett. 3 (2018) 2620–2640. doi: 10.1021/acsenergylett.8b01552

  9. [9]

     B. Tang, L. Shan, S. Liang, et al., Energy Environ. Sci. 12 (2019) 3288–3304. doi: 10.1039/C9EE02526J

  10. [10]

      X. Jia, C. Liu, Z.G. Neale, et al., Chem. Rev. 120 (2020) 7795–7866. doi: 10.1021/acs.chemrev.9b00628

  11. [11]

      K.W. Nam, H. Kim, J.H. Choic, et al., Energy Environ. Sci. 12 (2019) 1999–2009. doi: 10.1039/C9EE00718K

  12. [12]

      L. Su, L. Liu, Y. Wang, et al., Chin. Chem. Lett. 31 (2020) 2358–2364. doi: 10.1016/j.cclet.2020.03.014

  13. [13]

      C. Xu, B. Li, H. Du, et al., Angew. Chem. Int. Ed. 51 (2012) 933–935. doi: 10.1002/anie.201106307

  14. [14]

      M.H. Alfaruqi, V. Mathew, J. Gim, et al., Chem. Mater. 27 (2015) 3609–3620. doi: 10.1021/cm504717p

  15. [15]

      Z. Hou, M. Dong, Y. Xiong, et al., Small (2020) 2001228. doi: 10.1002/smll.202001228

  16. [16]

      H. Pan, Y. Shao, P. Yan, et al., Nat. Energy 1 (2016) 16039. doi: 10.1038/nenergy.2016.39

  17. [17]

      W. Sun, F. Wang, S. Hou, et al., J. Am. Chem. Soc. 139 (2017) 9775–9778. doi: 10.1021/jacs.7b04471

  18. [18]

      J. Wang, J.G. Wang, H. Liu, et al., Adv. Funct. Mater. 31 (2020) 2007397.

  19. [19]

      X.Z. Zhai, J. Qu, J. Wang, et al., Energy Storage Mater. 42 (2021) 753–763. doi: 10.1016/j.ensm.2021.08.021

  20. [20]

      X. Guo, J. Zhou, C. Bai, et al., Mater. Today Energy 16 (2020) 100396. doi: 10.1016/j.mtener.2020.100396

  21. [21]

      B. Lee, H.R. Seo, H.R. Lee, et al., ChemSusChem 9 (2016) 2948–2956. doi: 10.1002/cssc.201600702

  22. [22]

      J. Yang, J. Cao, Y. Peng, et al., ChemSusChem 13 (2020) 4103. doi: 10.1002/cssc.202001216

  23. [23]

      N. Zhang, F. Cheng, J. Liu, et al., Nat. Commun. 8 (2017) 405–415. doi: 10.1038/s41467-017-00467-x

  24. [24]

      S. Lian, C. Sun, W. Xu, et al., Nano Energy 62 (2019) 79–84. doi: 10.1016/j.nanoen.2019.04.038

  25. [25]

      D. Wang, L. Wang, G. Liang, et al., ACS Nano 13 (2019) 10643–10652. doi: 10.1021/acsnano.9b04916

  26. [26]

      G. Fang, C. Zhu, M. Chen, et al., Adv. Funct. Mater. 29 (2019) 1808375. doi: 10.1002/adfm.201808375

  27. [27]

      J. Huang, Z. Wang, M. Hou, et al., Nat. Commun. 9 (2018) 2906. doi: 10.1038/s41467-018-04949-4

  28. [28]

      J. Wang, J.G. Wang, X. Qin, et al., ACS Appl. Mater. Interfaces 12 (2020) 34949–34958. doi: 10.1021/acsami.0c08812

  29. [29]

      T. Zhang, Y. Tang, G. Fang, et al., Adv. Funct. Mater. (2020) 2002711. doi: 10.1002/adfm.202002711

  30. [30]

      S. Zhao, B. Han, D. Zhang, et al., J. Mater. Chem. A 6 (2018) 5733–5739. doi: 10.1039/C8TA01031E

  31. [31]

      B. Lee, H.R. Lee, H. Kim, et al., Chem. Commun. 51 (2015) 9265–9268. doi: 10.1039/C5CC02585K

  32. [32]

      J. Wang, J.G. Wang, H. Liu, et al., J. Mater. Chem. A 7 (2019) 13727–13735. doi: 10.1039/C9TA03541A

  33. [33]

      X. Zhu, Z. Cao, W. Wang, et al., ACS Nano 15 (2021) 2971–2983. doi: 10.1021/acsnano.0c09205

  34. [34]

      X.Z. Zhai, J. Qu, S.M. Hao, et al., Nano Micro Lett. 12 (2020) 56. doi: 10.1007/s40820-020-0397-3

  35. [35]

      T. Sun, Q. Nian, S. Zheng, et al., Small 16 (2020) 2000597. doi: 10.1002/smll.202000597

  36. [36]

      M.H. Alfaruqi, J. Gim, S. Kim, et al., J. Power Sources 288 (2015) 320–327. doi: 10.1016/j.jpowsour.2015.04.140

  37. [37]

      Y. Zeng, X. Zhang, Y. Meng, et al., Adv. Mater. 29 (2017) 1700274. doi: 10.1002/adma.201700274

  38. [38]

      M. Toupin, T. Brousse, D. Bélanger, Chem. Mater. 16 (2004) 3184–3190. doi: 10.1021/cm049649j

  39. [39]

      Y. Hou, Y. Cheng, T. Hobson, et al., Nano Lett. 10 (2010) 2727–2733. doi: 10.1021/nl101723g

  40. [40]

      G. Fang, J. Zhou, A. Pan, et al., ACS Energy Lett. 3 (2018) 2480–2501. doi: 10.1021/acsenergylett.8b01426

  41. [41]

      Y. Huang, J. Mou, W. Liu, et al., Nano Micro Lett. 11 (2019) 49–56. doi: 10.1007/s40820-019-0278-9

  42. [42]

      Y. Zhong, X. Xu, J. Veder, et al., iScience 23 (2020) 100943. doi: 10.1016/j.isci.2020.100943

  43. [43]

      S.H. Kim, S.M. Oh, J. Power Sources 72 (1998) 150–158. doi: 10.1016/S0378-7753(97)02703-1

  44. [44]

      Q. Chen, J. Jin, Z. Kou, et al., Small 16 (2020) 2000091. doi: 10.1002/smll.202000091

  45. [45]

      F. Kataoka, T. Ishida, K. Nagita, et al., ACS Appl. Energy Mater. 3 (2020) 4720–4726. doi: 10.1021/acsaem.0c00357

  46. [46]

      J. Huang, X. Tang, K. Liu, et al., Mater. Today Energy 17 (2020) 100475. doi: 10.1016/j.mtener.2020.100475

  47. [47]

      X. Gao, H. Wu, W. Li, et al., Small 16 (2020) 1905842. doi: 10.1002/smll.201905842

  48. [48]

      S.D. Han, S. Kim, D. Li, et al., Chem. Mater. 29 (2017) 4874–4884. doi: 10.1021/acs.chemmater.7b00852

  49. [49]

      L. Ma, M.A. Schroeder, O. Borodin, et al., Nat. Energy 5 (2020) 743–749. doi: 10.1038/s41560-020-0674-x

  50. [50]

      X. Han, H. Leng, Y. Qi, et al., Chem. Eng. J. 431 (2022) 133931. doi: 10.1016/j.cej.2021.133931

  51. [51]

      J. Hao, B. Li, X. Li, et al., Adv. Mater. (2020) 2003021. doi: 10.1002/adma.202003021

  52. [52]

      L. Wang, G. Fan, J. Liu, et al., Chin. Chem. Lett. 32 (2021) 1095–1110. doi: 10.1016/j.cclet.2020.08.022

• 

  Figure 1  (a) XRD pattern, (b) SEM, (c) TEM, SAED (inset) and (d) HRTEM images of MnO₂ sample.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) CV curves, (b) galvanostatic charge/discharge profiles of MnO₂ cathodes in 2 mol/L ZnSO₄ + 0.1 mol/L MnSO₄ electrolyte. Cycling performance of cells at a current rate of (c) 0.1 A/g and (e) 1 A/g (200 µL electrolyte). (d) Effect of electrolyte amount on the specific capacity.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) The initial two charge/discharge profiles, (b) ex-situ XRD patterns of the MnO₂ cathodes collected from different states. SEM, TEM, SAED (inset), HRTEM, and EDX elemental mapping images of the cathodes at (c) pristine A state, (d–f) fully-discharged B state, (g–i) fully-charged C state, and (j–l) partially-discharged D state.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) EIS spectra of the fully-discharged cathodes before and after acid washing. (b) Charge/discharge curves of the cells re-assembled with the acid-washed cathodes at 0.05 A/g. (c) Charge/discharge curves of the cells re-assembled with the discharged cathodes without acid washing. (d) Schematic illustration of working mechanism of α-MnO₂ cathodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) XRD pattern and (b) SEM image of the MnO₂ cathodes after the cycling test at 0.1 A/g. (c) Cycling performance of the cycled MnO₂ cathode re-assembled with a fresh Zn anode, and the inset shows the corresponding charge/discharge curves. (d) XRD of the Zn anodes in different electrolytes. (e–g) SEM images of the Zn anodes after soaking in different electrolytes for 10 days. (h) Long-term cycling stability of the Zn/Zn symmetric cells in different electrolytes.

  下载: 全尺寸图片 幻灯片
•