Unveiling the origin of the capacity fade in MnO$_{2}$ zinc-ion battery cathodes through an analysis of the Mn vacancy formation

TL;DR

This study addresses the Mn dissolution–driven capacity fade in MnO$_2$ cathodes for rechargeable aqueous Zn–ion batteries by computing Mn vacancy formation energies $E_d$ in both charged MnO$_2$ and discharged ZnMn$_2$O$_4$ phases across α- and λ-MnO$_2$ polymorphs, using DFT with defect-correction schemes. It demonstrates that Mn vacancy formation is energetically unfavorable in MnO$_2$ but becomes feasible in α-ZnMn$_2$O$_4$ due to unstable Zn coordination, with large relaxation-driven energy gains that stabilize Zn at the vacancy site. Operando $^1$H NMR corroborates the theory by showing continuous Mn$^{2+}_{(aq)}$ dissolution during discharge and partial redeposition upon charging, indicating a dissolution mechanism linked to Zn coordination rather than a Jahn–Teller–assisted disproportionation. The work suggests mitigation strategies via foreign-atom doping or Mn-site alloying to raise $E_d$, and argues that the identified defect-coordination mechanism may apply to other intercalation cathodes, guiding design toward longer-lived, grid-scale Zn–ion batteries.

Abstract

Currently explored rechargeable aqueous zinc-ion battery (RAZIB) cathode materials, such as $α$-MnO$_{2}$, suffer from severe capacity fade when cycling at rates appropriate for grid-scale operation. Mn dissolution has been previously identified as the cause of $α$-MnO$_{2}$ cathode degradation during RAZIB cycling, with conflicting evidence being found in support of the proposed Jahn-Teller effect-assisted charge disproportionation reaction as the mechanism behind Mn dissolution. In order to unveil the Mn dissolution mechanism in MnO$_{2}$ cathode cells under RAZIB operation conditions, the energetic feasibility for Mn vacancy formation was probed in both charged (MnO$_{2}$) and discharged (ZnMn$_{2}$O$_{4}$) phases of $α$ and $λ$ polymorphs of MnO$_{2}$ using density functional theory. The formation of a Mn vacancy, and consequently the dissolution of Mn as Mn$^{2+}_{(aq)}$, was found to be thermodynamically feasible for the $α$-ZnMn$_{2}$O$_{4}$ phase due to the energetically unfavourable Zn bent coordination formed during the Zn$^{2+}$ intercalation process, indicating that Mn dissolution is promoted by an unstable Zn coordination environment. The theoretical calculations were then corroborated by operando $^{1}$H nuclear magnetic resonance experiments which captured the Mn dissolution occurring throughout the RAZIB discharge, with subsequent electrochemical deposition of the Mn atoms on the electrode during charge. The combined computational and experimental analysis reveals the critical role of defect energetics and coordination environment in driving active material dissolution, and consequently capacity fade, with the proposed mechanism also relevant for understanding cathode degradation in other intercalating ion battery chemistries.

Figures (16)

• Figure 1: Relaxed crystal structures of bulk (a) $\alpha$-MnO2 and (b) $\alpha$-ZnMn2O4. As described in the text, the $\alpha$-ZnMn2O4 Mn atoms have been classified as Mn1 (olive) or Mn2 (teal) depending on the orientation of their MnO6 octahedral with respect to the bent coordination of the Zn atoms.
• Figure 2: (a) VMn1 and VMn2 sites position in the $\alpha$-ZnMn2O4 crystal structure indicated by lighter shaded Mn1 and Mn2 atoms, respectively. Relaxed structures of (b) $\alpha$-ZnMn2O4(VMn1) and (c) $\alpha$-ZnMn2O4(VMn2) for neutral charge defects ($q~=~0$), with the ZnO6 octahedra established after relaxation highlighted as a polyhedron. Only a zoomed-in region of the respective simulated $\alpha$-ZnMn2O4(VMn) cells is presented here to facilitate the visual analysis of the VMn positioning and atomic displacements during the relaxation of the Mn vacant structures.
• Figure 3: (a,c) $\alpha$-ZnMn2O4(VMn1) and (b,d) $\alpha$-ZnMn2O4(VMn2) structures (a,b) before and (c,d) after atomic relaxation, featuring the regions of locpot $>$ 10 eV in each structure.
• Figure 4: The EdVMn0 for the electrode calculated with respect to the $\alpha$-MnO2 cathode GCD experiment result (initial active material loading equal to 4.5 mg). The experimental GCD cycling potential profile used for calculating EdVMn0 is also shown in the figure on a secondary axis. The green and red regions highlight the EdVMn0 results for which the Mn dissolution is respectively thermodynamically unfavourable (EdVMn0 $>$ 0 eV) and favourable (EdVMn0 $<$ 0 eV). The specific capacity shown in the figure is the accumulated specific capacity expended and restored through the discharge and charge cycles. The arrows present in the graph highlight the positions in the EdVMn0 curve for the electrode where the $\alpha$-MnO2, $\alpha$-Zn_0.5Mn2O4, and $\alpha$-ZnMn2O4 phases are considered to be constituting the electrode.
• Figure 5: (a) Contour plot for chemical shift peak intensity during operando1HNMR experiment of a RAZIB utilizing an $\alpha$-MnO2 cathode, with its corresponding voltage curve during cycles 2 to 4. (b) Gravimetric capacity and associated Mn dissolved percentage with respect to the initial mass in the electrode for the discharged and charged states of the RAZIB cell via in situ1HNMR.