Zinc-air batteries have a significant promise for next-generation energy storage due to their low cost, high safety, high theoretical capacity, and abundant availability. Recently, we reported a new rechargeable zinc-air battery based on zinc peroxide chemistry. Compared to conventional zinc-air batteries, this neutral zinc-air battery exhibits higher reversibility and stability, making it a more promising candidate for secondary zinc-air batteries. However, the low energy efficiency of neutral zinc-air batteries remains a challenge. And the mechanism of these neutral zinc-air battery reactions is still unclear. In this work, we systematically investigated the reaction pathways of neutral zinc-air reaction by first-principles calculations.

Density Functional Theory calculations based on a software package named VASP was employed with PAW potentials and the PBE functional. The vdW interactions were corrected by the Grimme's D3 method. The k-point mesh was generated by the Monkhorst-Pack scheme via Pymatgen with MPRelaxSet settings. A cutoff energy of 520 eV was utilized within a convergence criterion of 10^-5 eV for self-consistent calculations. The convergence criterion for ionic relaxation was set to be 0.02 eV/Å. The calculations were performed by a slab model with a thickness of approximately 10 Å and a vacuum layer of greater than 15 Å.

There is a clear linear scaling relationship among the adsorption strengths of O*, ZnO₂*, and ZnO₄* in neutral zinc-air reactions. This relationship constrains the freedom in catalyst design, making it difficult to optimize all intermediate adsorption strengths simultaneously. Breaking this linear scaling relationship will be a focus of future research. The rate-determined step for both the four-electron and two-electron reactions in Ag (111) is the first electron transfer. A weaker ZnO₄* adsorption makes the first reaction barrier as high as 0.78 eV at 1.2 V (vs. Zn/Zn²⁺). This result indicates that even with this metal catalyst, the discharge voltage of neutral zinc-air batteries remains smaller than that of alkaline batteries due to the limitation of linear scaling relationships.

Among various catalysts, Pt (111) exhibits a lowest reaction barrier, indicating that Pt maintains a high catalytic activity in Zn²⁺-involved ORR processes. Conversely, the surface reactivity of Ru (111) and Co (001) catalyst surfaces is significantly greater than that of Ag (111), where over-strong ZnO₄* adsorption makes the second electron transfer the rate-determined step. Meanwhile, the energy barrier of four-electron pathway is notably lower than that of two-electron transfer pathway, showing that the Ru (111) and Co (001) catalyst can favor a four-electron transfer reaction and form ZnO discharge products.

The optimal O* adsorption free energy is found to be around -6 eV based on the linear scaling relationships. And Ag (111) and Pt (111) have relatively appropriate O* adsorption strengths. The neutral zinc-air catalytic activity of Ag and Pt can be further improved via increasing the d-band center. In addition, the calculations also reveal that catalysts with a strong reactivity favor a four-electron transfer pathway.

The theoretical calculation results showed that linear scaling relationships still had true among various adsorption intermediates in neutral zinc-air reactions. Oxygen adsorption strength and overpotential followed the Sabatier principle, indicating that a moderate strength (i.e., ΔG_O* ~ -6 eV) yielded an optimal catalytic activity. Among numerous catalysts, Ag and Pt exhibited an optimal surface reactivity with a high theoretical catalytic activity. Furthermore, employing catalysts with a strong surface reactivity, such as Co and Ru, represented a potential approach to achieve non-protonic four-electron zinc-air batteries.