Understanding the active site in chameleon-like bifunctional catalyst for practical rechargeable zinc-air batteries | Nature Communications

Nature Communications volume 15, Article number: 9616 (2024) Cite this article

Metrics details

The practical application of rechargeable zinc-air batteries faces challenges stemming from inadequate bifunctional catalysts, contradictory gas-liquid-solid three-phase interfaces, and an ambiguous fundamental understanding. Herein, we propose a chameleon-like bifunctional catalyst comprising ruthenium single-atoms grafted onto nickel-iron layer double hydroxide (RuSA-NiFe LDH). The adaptive oxidation of RuSA-NiFe LDH to oxyhydroxide species (RuSA-NiFeOOH) during charging exposes active sites for the oxygen evolution reaction, while reversible reduction to NiFe LDH during discharge exposes active sites for the oxygen reduction reaction. Additionally, a hierarchical air cathode featuring hydrophilic and hydrophobic layers facilitates the reversible conversion between RuSA-NiFe LDH and RuSA-NiFeOOH, expedites oxygen bubble desorption, and suppresses carbon corrosion. Consequently, our zinc-air batteries demonstrate a high charge/discharge capacity of 100 mAh cm−2 per cycle, a voltage gap of 0.67 V, and an extended cycle life of 2400 h at 10 mA cm−2. We comprehensively elucidate the catalytic reaction thermodynamics and kinetics for the air cathode through electrode potential decoupling monitoring, oxygen bubble desorption tracking, and carbon content quantification.

The pressing need to combat global warming and achiev carbon neutrality has intensified the focus on in developing sustainable rechargeable batteries for large-scale energy storage1,2,3,4. Among various battery technologies, rechargeable zinc-air batteries (RZABs) stand out as promising candidates, offering an impressive theoretical specific energy of ~1370 Wh kg−1 based on the weight of Zn, 2Zn + O2 = 2ZnO. Additionally, RZABs have the advantages of inherent safety (aqueous electrolyte), low cost (zinc is ~3200 US dollars per ton) and environmental friendliness (non-toxic chemicals and mature recycling technology for zinc)5,6,7,8. However, their practical application is limited by the poor activity and oxidation vulnerability of bifunctional catalysts at high charge voltages, as well as challenges in the gas-liquid-solid three-phase interfaces in the air cathode9,10,11. Consequently, the actual specific energy falls short of theoretical values, with low charge/discharge capacity (less than 5 mAh cm−2) per cycle, severe polarization (charge/discharge voltage differences larger than 0.8 V), and short cycle lifespan (less than 500 h)12,13,14. Furthermore, the lack of effective characterization tools has impeded a thorough understanding of side reactions and voltage fluctuations at the air cathode15. To overcome these limitations, two key scientific challenges must be addressed: regulating active sites in bifunctional catalysts to enhance catalytic activity and stability, and clearly elucidating the mechanisms of oxygen bubble evolution, side reactions, and voltage fluctuations at the air cathode16.

Monofunctional catalysts for the oxygen evolution reaction (OER) or oxygen reduction reaction (ORR) have been extensively studied and applied due to their good catalytic performance17,18,19. However, bifunctional OER/ORR catalysts often suffer from poor activity and stability, as many ORR active sites become oxidized and deactivated at high potentials20. Designing a catalyst that can adaptively regulate its active sites in response to the redox environment is therefore crucial for achieving high stability and activity21. Transition metal hydroxides are known for their outstanding OER stability and activity, but exhibit poor ORR performance22. In contrast, single-atom catalysts (SACs) offer superior catalytic performance with their maximum metal utilization, uniform active sites, and tunable designability23,24. The M-O bond in the SACs (where M is a transition metal) serves as the active site, providing good ORR activity and good stability under oxidation conditions25,26. The ordered crystal can trap single metal atoms through a strong covalent metal-support interaction, improving the stability of SACs27,28,29. Incorporating transition metal single atoms into metal hydroxides, which have abundant oxygen sites and vacancies30,31,32, is an effective strategy for developing adaptive bifunctional catalysts with enhanced OER/ORR activity and stability.

In addition to adaptive catalyst design, the construction of the air cathode is vital for the diffusion of oxygen bubbles and hydroxide ions (OH−), impacting catalyst utilization, reaction kinetics, and overall RZAB performance33. Understanding the three-phase interfacial reactions and the role of air cathode in RZABs remains a challenge and is increasingly seen as a bottleneck in improving performance. The gradual increase in the charge/discharge voltage gap in RZABs has traditionally been attributed to bifunctional catalyst degradation, neglecting the effects of oxygen bubble desorption and conductive carbon oxidation at the air cathode34. Decoupling the cathode potential from the battery voltage is critical to understanding and mitigating the large charge/discharge voltage gap in RZABs. Typically, air cathodes are fabricated with hydrophobic binders to maintain a hydrophobic three-phase interface suitable for the discharge process (O2 + 2H2O + 4e− → 4OH−), which prevents flooding35. However, the charging process (4OH− → O2 + 2H2O + 4e−) requires hydrophilic and superaerophobic interfaces to manage oxygen bubble adhesion effectively. Traditional air cathodes struggle with severe oxygen bubble adhesion during charging, leading to dead zones, low catalyst utilization, high ohmic resistance, a large charge/discharge voltage gap, and uneven zinc deposition36,37. A comprehensive air cathode evaluation system is essential to unlock this “black box”, elucidating the degradation mechanisms and guiding air cathode design.

In this work, ruthenium single atoms grafted onto nickel-iron layered double hydroxide (RuSA-NiFe LDH) were screened from eleven transition metals and optimized using density functional theory (DFT) calculations. This catalyst demonstrates the ability to reversibly tailor the thermodynamic active sites between RuSA-NiFe LDH and RuSA-NiFeOOH during ORR and OER, behaving like a chameleon. Thanks to this self-adaptive behavior, the RuSA-NiFe LDH exhibits a small OER/ORR potential gap (∆E = Ej=10 − E1/2) of just 0.554 V. We propose an innovative hierarchical air cathode (RuSA-NiFe LDH HE) featuring hydrophilic and hydrophobic layers, which facilitates the reversible conversion between RuSA-NiFe LDH and RuSA-NiFeOOH, accelerates oxygen bubble desorption, lowers charge voltage, and mitigates conductive carbon oxidation. Leveraging RuSA-NiFe LDH HE, the resulting RZABs achieve remarkable charge/discharge capacity of 100 mAh cm−2 per cycle and an extended cycle life of 2400 h, surpassing the most reported RZABs. To demonstrate the practical applicability, a scaled-up RZAB pack was integrated with solar panels to store energy, and an ampere-hour level Zn-air pouch cell with high specific energy was also developed. This work not only presents an efficient chameleon-like strategy for enhancing RZAB performance through bifunctional catalyst design and air cathode construction but also introduces in-situ techniques to unravel the roles of bifunctional catalysts and air cathodes in RZABs.

The NiFe LDH transforms into NiFeOOH under oxidative conditions, displaying heightened OER activity and stability. Subsequently, NiFeOOH can be reverted to NiFe LDH under reducing potential. However, NiFe LDH shows poor ORR activity38. Therefore, fine-tuning the active site of NiFe LDH to enhance ORR performance is an efficient strategy for acquiring a bifunctional OER/ORR catalyst (Fig. 1a). The metal single-atom active sites have high ORR activity and stability39. The NiFe LDH has abundant functional groups to trap metal single atoms. Thirteen transition metal single-atoms (V, Cr, Mn, Co, Cu, Zn, Mo, Ru, Rh, Pd, Ag, Ce and Pt) were used and introduced on the NiFe LDH (Fig. 1b). We investigate the Gibbs free energy change during the reaction pathways to identify the model with the best OER and ORR intrinsic activities (Supplementary Figs. 1–12 and Table 1). Because the surface of NiFe LDH would be partially oxidized to form NiFeOOH during the OER process, the surface of NiFe LDH is replaced by NiFeOOH for OER calculation (Supplementary Data 1)40. Among these single-atom introductions, the RuSA-NiFeOOH and RuSA-NiFe LDH have the lowest free energy changes of the rate-determining step (RDS) of 0.34 eV for OER (Supplementary Fig. 1a) and 0.41 eV for ORR (Supplementary Fig. 7a), respectively. The volcano plots (Fig. 1c) show that the NiFeOOH has comparable OER activity; however, NiFe LDH has too strong *OH binding for *OH to desorb, which leads to poor ORR activity. After introducing the Ru single-atom, it has an optimal adsorption-free energy difference between *O and *OH for OER, which favors forming *OOH. The Ru single-atom has an optimal adsorption-free energy of *OH for ORR, which prefers to *OH desorption. The density functional theory (DFT) calculations reveal that the introduction of Ru single-atom exhibits higher OER/ORR catalytic activities than other transition metal single-atoms.

a Schematic representation of a bifunctional OER/ORR catalyst design based on NiFe LDH. b Screen transition metal single-atoms on the NiFe LDH. c Volcano plots of the OER theoretical overpotential (ηOER) versus adsorption free energy difference (∆G*O–∆G*OH) and ORR theoretical overpotential (ηORR) versus adsorption free energy (∆G*OH). d Atom-resolution HAADF-STEM image of RuSA-NiFe LDH (Ru single atoms marked as orange circles). e Zoomed-in intensity profile of the Ru single-atom and the surrounding Ni (Fe) atoms. f The line profile for HAADF intensity analysis is labeled in (d). g EDS elemental mappings of Ni, Fe, and Ru. h–j Ru K-edge Normalized XANES (h), Fourier transformed EXAFS spectra (i) in R space, and WT EXAFS spectra (j) of the RuSA-NiFe LDH, RuO2, and Ru foil.

Guided by the theoretical results, we chose the model of Ru single-atoms decorated on NiFe LDH substrate, which was prepared via a one-step hydrothermal reaction (Supplementary Fig. 13). The as-prepared sample was collected by filtrating, washing, and freeze-drying. The X-ray diffraction (XRD) pattern shows that the RuSA-NiFe LDH has the same crystalline phase as NiFe LDH (Supplementary Fig. 14, JCPDS No. 51-0463). Increasing the amount of the RuCl3·xH2O, the RuO2 nanoparticles are formed, and the sample is denoted as RuNP-NiFe LDH (Supplementary Fig. 14). The scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images, along with XPS results (Supplementary Figs. 15–20) show the NiFe LDH and RuSA-NiFe LDH with a nanosheet morphology, and no nanoparticles are found on the nanosheet, but some RuO2 nanoparticles were observed on the RuNP-NiFe LDH nanosheet. The inductively coupled plasma-optical emission spectrometry (ICP-OES) was carried out to quantify the contents of metal, showing the mass ratio of Ni:Fe is close to 3:1 (Supplementary Fig. 21a). The Ru contents in the RuSA-NiFe LDH and RuNP-NiFe LDH are 0.28 wt% and 7.34 wt% (Supplementary Fig. 21b), respectively. The Brunaer-Emmett-Teller (BET) specific surface areas of RuSA-NiFe LDH, RuNP-NiFe LDH, and NiFe LDH are 53.0, 41.7, and 43.1 m2 g−1 (Supplementary Fig. 21c), respectively, which reveals RuSA-NiFe LDH exposing more active sites. Aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image (Fig. 1d) shows an amount of intense bright spots, which are regarded as Ru single-atoms due to their higher reflection electronic intensity than the surrounding Ni(Fe) atoms. The intensity profiles (Fig. 1e, f) show a higher intensity site of Ru atoms than that of Ni(Fe) atoms, which also reveals the isolated Ru single-atoms anchored on the surface of NiFe LDH. The X-ray energy-dispersive spectroscopy (EDS) mappings (Fig. 1g) show that the Ni and Fe are uniformly dispersed at the nanosheet, and the Ru single-atoms are individually dispersed on the nanosheet.

X-ray absorption spectroscopy (XAS) was employed to study the chemical state further and coordinate the environment of Ru atoms in the RuSA-NiFe LDH. The X-ray absorption near-edge structure (XANES) spectra of Ru K-edge show that the white line intensity of RuSA-NiFe LDH is between that of Ru foil and RuO2 (Fig. 1h), which reveals the valence state of Ru atoms in the RuSA-NiFe LDH is between 0 and +4. The linear fitting relation between the oxidation state and pre-edge location shows that the oxidation state of Ru is around +1.9 in the RuSA-NiFe LDH. The Ru K-edge extended X-ray absorption near-edge fine structure (EXAFS) spectra show that RuSA-NiFe LDH has only one well-defined coordination shell at around 1.5 Å (Fig. 1i), which is attributed to the scattering of the Ru-O bond. The peak at 2 ~ 3 Å shows a coordination number of 1.1 ± 0.3 (Supplementary Fig. 22 and Table 2), and its low intensity suggests that the few Ru single atoms maybe coordinated with surrounding Ni or Fe atoms. No strong peak of Ru-Ru or Ru-O-Ru bonds has been found at 2–3 Å, which reveals the atomic dispersion of single Ru atoms and no formation of Ru cluster in the RuSA-NiFe LDH41. The wavelet transform (WT) analysis shows only one intensity maximum in RuSA-NiFe LDH (Fig. 1j), confirming the Ru dispersed atomically on the NiFe LDH substrate. The EXAFS fitting results also confirm the coordination of Ru single atoms with oxygen.

The electrocatalytic performances for the OER and ORR were thoroughly evaluated in a 1.0 and 0.1 mol L−1 KOH aqueous electrolytes, respectively. All potentials were referenced against the reversible hydrogen electrode (RHE) using a calibrated Hg/HgO reference electrode (Supplementary Fig. 23). A free-standing electrode was prepared using carbon felt (CF) as the substrate and the catalysts in-situ growth on the CF during the hydrothermal process. All polarization curves for the free-standing electrodes were recorded from high to low potential to mitigate the interference from the oxidation peak of NiFe(OH)2/NiFeOOH42. The RuSA-NiFe LDH/CF has a potential (Ej=10) of 1.446 V at a current density of 10 mA cm−2 (Fig. 2a), surpassing RuNP-NiFe LDH/CF (1.50 V), NiFe LDH/CF (1.46 V), IrO2/C (1.56 V) and CF (1.67 V). Tafel plots, calculated from the linear sweep voltammetry (LSV) curves, reveal the lowest Tafel slope of 24.8 mV dec−1 for RuSA-NiFe LDH/CF among these catalysts (Supplementary Fig. 24a), signifying rapid OER kinetics. The OER activities of powder samples on a rotating disk electrode (RDE) were also evaluated. The RuSA-NiFe LDH demonstrates a lower Ej=10 of 1.46 V and a lower Tafel slope of 56.7 mV dec−1 (Supplementary Fig. 24b, c) compared to RuNP-NiFe LDH (1.54 V, 128.5 mV dec−1), NiFe LDH (1.50 V, 109.1 mV dec−1) and IrO2/C (1.59 V, 109.9 mV dec−1). These results indicate the high OER catalytic activity of RuSA-NiFe LDH, whether in powder form or on a conductive substrate. The ORR performances of all powder catalysts were evaluated using an RDE. RuSA-NiFe LDH exhibits a half-wave potential (E1/2) of 0.892 V, superior to RuNP-NiFe LDH (0.80 V), NiFe LDH (0.58 V) and Pt/C (0.86 V) (Fig. 2b), showcasing high ORR activity. The RuSA-NiFe LDH has a high limiting current density of 6.48 mA cm−2, greater than other catalysts. The RuSA-NiFe LDH shows a current density increase of 1.2 mA cm−2 with increasing rotational speed at each step, larger than Pt/C (0.7 mA cm−2) (Supplementary Figs. 24d, e). This indicated that the RuSA-NiFe LDH exhibits rapid kinetics under adequate oxygen mass transfer43. The kinetics current densities (jk) at 0.85 V are 15.94 mA cm−2 for RuSA-NiFe LDH, 6.68 mA cm−2 for Pt/C, 0.93 mA cm−2 for RuNP-NiFe LDH and 0.01 mA cm−2 for NiFe LDH (Supplementary Fig. 24f), further highlighting the fast kinetics for RuSA-NiFe LDH. The i–t curves display that RuSA-NiFe LDH shows no noticeable degradation, while Pt/C exhibits a dramatic decrease after injecting 2 mL methanol (Supplementary Fig. 24g), indicating the outstanding methanol tolerance for RuSA-NiFe LDH. Rotating ring-disk electrode (RRDE) tests reveal an electron transfer number of 3.86 and a maximum HO2- yield of 13.1% for RuSA-NiFe LDH (Supplementary Fig. 24h, i), suggesting O2 conversion to OH− through a dominant four-electron transfer pathway. The original LSV curves for OER and ORR without iR compensation were shown in Supplementary Fig. 25. The durability of bifunctional catalysts is a crucial indicator of their practicality. RuSA-NiFe LDH maintains stable potential at various current densities of 10, 25, 50, and 100 mA cm−2. It also shows an long stability of 2000 h (Fig. 2c) without noticeable degradation. During ORR, RuSA-NiFe LDH retains 85.4% of its initial current at a constant potential of 0.7 V after 100 h, while the Pt/C decreases to 79.2% after 26 h (Fig. 2d). This underscores the good stability of RuSA-NiFe LDH for both OER and ORR, emphasizing the critical role of atomically dispersed Ru single-atoms for OER and ORR on the NiFe LDH substrate. Bifunctional catalysts with superior oxygen reversibility display a small potential gap (∆E = Ej=10 − E1/2) between the OER and ORR potentials, making ∆E a widely recognized indicator for the bifunctional activity44. A comparison of OER/ORR activities revealed that RuSA-NiFe LDH had an ∆E of 0.554 V, lower than Pt/C + IrO2/C and most previously reported bifunctional catalysts (Fig. 2e and Supplementary Table 3). These electrochemical results confirm the good OER/ORR activity and stability of RuSA-NiFe LDH simultaneously, highlighting its high suitability for RZABs.

a OER LSV curves with iR compensation of catalysts on Ni foam in 1.0 mol L−1 KOH electrolyte, where R was determined to be 0.3 ± 0.2 Ω. b ORR polarization curves with iR correction of catalysts in an O2-saturated 0.1 mol L−1 KOH electrolyte under a rotation speed of 1600 rpm, where R was determined to be 35 ± 4 Ω. c Chronopotentiometry curves of RuSA-NiFe LDH at the current densities from 10, 25, 50, 100 mA cm−2 and back to 10 mA cm−2. d Relative current of RuSA-NiFe LDH and 20 wt% Pt/C at 0.7 V (vs. RHE) with a rotation speed of 900 rpm. e Comparison of OER and ORR activities (Ej=10 and E1/2) of RuSA-NiFe LDH and other reported bifunctional catalysts. f Galvanostatic charge/discharge cycling curves of RZABs based on various catalysts at 10 mA cm−2. Each cycle takes 30 min to charge and 30 min to discharge. g In-situ Raman spectra I/Iν contour map of RuSA-NiFe LDH during charge/discharge processes and corresponding charge/discharge profiles at 10 mA cm−2. h Selected in-situ Raman spectra acquired during charge/discharge processes. i Ex-situ high-resolution Ni 2p XPS spectra of RuSA-NiFe LDH and corresponding Ni3+/Ni2+ contents at pristine state, after charge, and after discharge.

To further evaluate the high oxygen reversible activity and stability of RuSA-NiFe LDH in RZABs, home-made RZABs were assembled with RuSA-NiFe LDH air cathode, Zn plate, and alkaline electrolyte (6 mol L−1 KOH with 0.2 mol L−1 Zn(Ac)2·2H2O and saturated ZnO). RZABs with the RuSA-NiFe LDH deliver an extraordinary long cycle life of 2100 h and a small voltage gap of 0.702 V at 10 mA cm−2 (Fig. 2f). In comparison, the NiFe LDH air cathode has a cycle life of 1057 h and a voltage gap of 0.849 V, and the Pt/C + IrO2/C air cathode shows a cycle life and voltage gap of 325 h and 0.932 V, respectively. The electrochemical performance of RZABs based on RuSA-NiFe LDH greatly surpasses those of RZABs with most of the reported bifunctional catalysts. To gain deeper insights into the catalytic mechanism of RuSA-NiFe LDH, which exhibits high oxygen reversible stability, we conducted in-situ Raman and ex-situ XPS experiments. The custome-built device for the in-situ Raman tests is shown in Supplementary Fig. 26. In the RuSA-NiFe LDH powder sample and RuSA-NiFe LDH electrode, four characteristic peaks were observed at 149.7, 292.6, 456.2 and 524.3 cm−1 (Supplementary Fig. 27), corresponding to the vibrations of Fe(Ni)-OH, Fe-O, δ(Ni-O) and ν(Ni-O)45. After the electrolyte was added to the cell, only two characteristic peaks, corresponding to the stretching modes of δ(Ni-O) and ν(Ni-O), were observed. To study the evolution of active species in RuSA-NiFe LDH, we obtained in-situ Raman spectra while cycling the cell at charge/discharge current densities of 10 mA cm−2. During the discharge process, the intensity of the δ(Ni-O) and ν(Ni-O) peaks showed no noticeable change (Fig. 2g, h), indicating that the surface of RuSA-NiFe LDH did not undergo a phase transition during the initial discharge stage. In the subsequent charge process, the intensity of δ(Ni-O) increased, while that of ν(Ni-O) decreased. The intensity ratio (Iδ/Iν) of δ(Ni-O) to ν(Ni-O) reached a maximum value after 30 min of charging, demonstrating the transformation of RuSA-NiFe LDH into RuSA-NiFeOOH at high oxidation potential46. When the cell returned to the discharge process, the Iδ/Iν ratio gradually decreased to its initial state (Fig. 2g, h), revealing the reversion of RuSA-NiFeOOH back to RuSA-NiFe LDH under a reduction current. Ex-situ XPS measurements were conducted to study the evolution of Ni and Fe in RuSA-NiFe LDH after the charging and discharging cycle. Initially, the pristine RuSA-NiFe LDH primarily contained Ni2+ species, with a binding energy of 854.7 eV (Ni2+ 2p3/2). After charging, the peak corresponding to Ni3+ 2p3/2 at 855.7 eV increased, with the Ni3+ content rising from 40.3% to 67.5% (Fig. 2i). This indicates that some Ni2+ species were oxidized to Ni3+, forming RuSA-NiFeOOH. Upon discharge, the intensity of the Ni3+ 2p3/2 peak decreased, and the Ni3+ content returned to 42.3%, suggesting the reduction of Ni3+ back to Ni2+. Furthermore, the pristine RuSA-NiFe LDH exhibits a Fe 2p3/2 peak at a binding energy of 711.3 eV (Supplementary Fig. 28). Upon charging, this peak shifts to 721.1 eV, indicating the oxidation of some Fe2+ species to Fe³⁺, forming RuSA-NiFeOOH. Both Ni and Fe act as active sites for the OER. During discharge, the Fe 2p3/2 peak shifts back to 711.8 eV, suggesting partial reduction of Fe3+ back to Fe2⁺. However, the reduction of Fe3+ is less siginificant compared to Ni3+, likely due to Ni3+ being more oxidizing than Fe3+. ICP-OES analysis reveals a Ni:Fe mass ratio close to 3:1, the relatively low Fe content allows us to exclude any unreduced Fe³⁺. Both in-situ Raman and ex-situ XPS confirmed the oxidation of RuSA-NiFe LDH to RuSA-NiFeOOH during the charging process, with RuSA-NiFeOOH serving as the catalytic site for the OER. During the subsequent discharge process, RuSA-NiFeOOH is reduced back to RuSA-NiFe LDH, which acts as the catalytic species for the ORR. These findings verify a new chameleon-like pathway for the reversible conversion between RuSA-NiFe LDH and RuSA-NiFeOOH during discharge/charge processes, enhancing stability and automatically tailoring active sites.

Our findings motivated us to propose a reversible pathway between OER and ORR. As Fig. 3a depicts, the RuSA-NiFe LDH is the active catalyst for ORR, and the RuSA-NiFe LDH is oxidized to RuSA-NiFeOOH when switching to the OER process. Returning to the ORR process, the RuSA-NiFeOOH can be reduced to RuSA-NiFe LDH, like a chameleon. In active structure, these reversible changes between RuSA-NiFe LDH and RuSA-NiFeOOH lead to high stability. The RuSA-NiFeOOH and RuSA-NiFe LDH show high OER (Supplementary Fig. 1a) and ORR (Supplementary Fig. 7a) catalytic activity, respectively, following the adsorbate evolution mechanism (AEM). This demonstrates a chameleon-like strategy to design a self-adaptive catalyst with high bifunctional activity and stability. To investigate the underlying origin of Ru introduction for superior OER and ORR activity further, the electronic structure features of metal moieties were studied. The RDS is over-strong adsorption of *O intermediate (*O + OH− → *OOH + e−) and *OH intermediate (*OH + e− → * + OH−) for OER and ORR, the charge density differences for adsorption O and OH for OER and ORR are studied, respectively. The adsorption of O in the NiFeOOH (Fig. 3b, c) and adsorption of OH in the NiFe LDH (Fig. 3d, e) show strong charge accumulation, which may lead to disfavoring of adsorbed OH- to form *OOH intermediate and the desorption of *OH to form OH−, respectively. After the Ru-single atom introduction, the adsorbed O in RuSA-NiFeOOH and adsorbed OH in the RuSA-NiFe LDH show suitable charge accumulation so that to form *OOH intermediate and to form OH- easily, respectively. The optimal adsorption energy for intermediates would decrease the energy barrier and improve the intrinsic activity.

a Schematic illustration of the reaction mechanism of RuSA-NiFeOOH for OER and RuSA-NiFe LDH for ORR in alkaline electrolyte, and the proposed reversible conversion of RuSA-NiFe LDH ↔ RuSA-NiFeOOH between OER and ORR process. b–e Charge density differences of O adsorption on RuSA-NiFeOOH (b) and NiFeOOH (c) with an isovalue of 0.01 | e|/Bohr−3, charge density differences of OH adsorption on RuSA-NiFe LDH (d) and NiFe LDH (e) with an isovalue of 0.01 |e|/Bohr−3, the charge accumulation and depletion regions are represented in yellow and blue, respectively. f Calculated surface Pourbaix diagram for the change from RuSA-NiFe LDH to RuSA-NiFeOOH and RuO3-NiFeO2. g–j The crystal orbital Hamilton population (COHP) of O from the adsorbed intermediates on the NiFeOOH (g), RuSA-NiFeOOH (h), NiFe LDH (i), RuSA-NiFe LDH (j), the integrated COHP values, expressed in eV per bond, are provided.

The Pourbaix diagram (Fig. 3f) shows that isolated Ru single-atom decorated on the surface of NiFe LDH in RuSA-NiFe LDH under low potential (Supplementary Fig. 30a–c). When the potential increases to 1.12 V (vs. SHE) at the pH of 14, partial NiFe LDH is oxidized to NiFeOOH, and the Ru single atom decorates on the surface of NiFeOOH to form RuSA-NiFeOOH. When the potential reaches a high value of 4.59 V (vs. SHE) at the pH of 14, NiFeOOH oxidizes to NiFeO2. Simultaneously, the Ru single atom is oxidized to RuO3 and detaches from the NiFeO2, which leads to poor OER activity. This Pourbaix diagram demonstrates that RuSA-NiFe LDH/RuSA-NiFeOOH can be reversed by controlling the oxidation/reduction potential, and the RuSA-NiFeOOH has a strong oxidation resistance and good stability. The projected density of states (PDOS) and crystal orbital Hamiltonian population (COHP) were studied to probe further the crucial role of Ru introduction in the OER/ORR activity. As shown in Supplementary Fig. 30a, b, the RuSA-NiFeOOH offers a more negative value (−2.24 eV) of the Ru d-band center than that of the Fe d-band center (−1.93 eV) in NiFeOOH, which would produce an increase of antibonding orbital filling and decrease the interaction between the active sites and the adsorbates. Moreover, the RuSA-NiFe LDH shows a more negative value (−2.98 eV) of the Ru d-band center than that of the Fe d-band center (−1.81 eV) in NiFe LDH (Supplementary Fig. 30c, d), which favors the *OH desorption to form OH−. A large negative COHP value indicates a strong bonding interaction47. The ICOHP values between O and Ru in RuSA-NiFeOOH and RuSA-NiFe LDH are more negative compared to those between O and Ni or Fe in NiFeOOH and NiFe LDH (Fig. 3g–j), respectively, indicating a stronger bond between Ru and O. The COHP diagrams reveal positive peaks of a certain height below the Fermi level, suggesting that the electrons of Ni, Fe and Ru may partially occupy the antibonding orbitals of O atoms. The height and width of these peaks suggesting this effect is more pronounced in RuSA-NiFeOOH and RuSA-NiFe LDH. Consequently, Ru single atoms weaken the bonding strength within the intermediates by filling the 2π* antibonding orbitals, thereby enhancing their catalytic activity48. DFT calculations confirm that the chameleon-like RuSA-NiFe LDH/RuSA-NiFeOOH has high activity and stability for RZABs based on conversion reactions.

Oxygen reversible reactions take place at the air cathode in RZABs, necessitating optimal gas-liquid-solid three-phase interfaces to enhance the conversion kinetics of active sites in the chameleon-like catalyst (Fig. 4a). This optimization maximizes active site utilization and facilitates the diffusions of oxygen bubbles/OH− ions, thereby enhancing overall performance. Unlike the traditional air cathode with a sole hydrophobic layer, we have developed a RuSA-NiFe LDH HE featuring an additional hydrophilic layer on top. This innovative design addresses the challenge of contradictory gas-liquid-solid three-phase interfaces during charge/discharge processes. The hydrophobic layer comprises RuSA-NiFe LDH powder, conductive carbon, and polytetrafluoroethylene (PTFE). RuSA-NiFe LDH/CF was used as a hydrophilic layer directly. The fabrication of RuSA-NiFe LDH HE utilized a roll-to-roll technology (Fig. 4b). Notably, both RuSA-NiFe LDH powder and RuSA-NiFe LDH/CF can be conveniently synthesized on a large scale through a hydrothermal step (Supplementary Fig. 31), demonstrating the potential for large-scale preparation of RuSA-NiFe LDH HE for commercial application. SEM images and EDS mappings illustrate that the RuSA-NiFe LDH nanosheet is vertically anchored into the CF, uniformly covering its surface (Supplementary Fig. 32), enhancing the electron transfer and augmenting the active surface area. Even after a prolonged OER test at 10 mA cm−2 for 2000 h, the RuSA-NiFe LDH nanosheets remained intact on the CF (Supplementary Fig. 33), indicating good structural stability of RuSA-NiFe LDH/CF. Side-view SEM image and EDS mappings indicate approximate thicknesses of 208 μm for the hydrophilic layer, 197 μm for the hydrophobic layer, and 330 μm for the diffusion layer, successively in the RuSA-NiFe LDH HE (Fig. 4c, d). These layers show close sequential contact, indicating low contact resistance and rapid electron transfer.

a Schematic oxygen bubbles desorption from hierarchical air cathode and a conventional air cathode. b Schematic illustrating the large-scale preparation of hierarchical air cathode for RZABs. c Cross-section SEM image of RuSA-NiFe LDH HE. d Corresponding EDS mappings of RuSA-NiFe LDH HE. e, f Electrolyte wetting ability testing of hydrophilic layer (e) and hydrophobic layer (f) in the RuSA-NiFe LDH HE. g, h Oxygen bubble contact angle of hydrophilic layer (g) and hydrophobic layer (h) in the RuSA-NiFe LDH HE. i, j Optical images of oxygen bubbles evolution behavior on the RuSA-NiFe LDH HE (i) and traditional RuSA-NiFe LDH air cathode (j) at 50 mA cm−2. k, l 3D CLSM images of Zn plate anodes after cycled in the RZABs for 300 h at 10 mA cm−2 with RuSA-NiFe LDH HE (k) and traditional RuSA-NiFe LDH air cathode (l). m, n Simulation of the effect of oxygen bubbles on electric field distribution in the electrolyte at the RuSA-NiFe LDH HE (m) and traditional RuSA-NiFe LDH air cathode (n) during the charging process. o, p Simulation of the effect of oxygen bubbles on electric field distribution at anode at the RuSA-NiFe LDH HE (o) and traditional RuSA-NiFe LDH air cathode (p) during the charging process.

The Ni mesh current collector embedded into the diffusion layer also facilitated swift electron transfer. In contrast, the traditional RuSA-NiFe LDH air cathode featured a hydrophobic layer with a thickness of approximately 200 μm (Supplementary Fig. 34). The hydrophilic layer is readily wetted upon electrolyte exposure (Fig. 4e), which indicates a super wettable interface with high affinity to the electrolyte. The hydrophobic layer exhibits a large contact angle of 145°, keeping for an extended duration (Fig. 4f), accelerating oxygen diffusion from the air during discharge and displaying strong flooding resistance. Furthermore, the hydrophilic layer shows a larger contact angle of 147° for oxygen bubbles compared to the hydrophobic layer (107°), suggesting that the hydrophilic layer facilitated faster detachment of oxygen bubbles due to lower adhesion (Fig. 4g, h). These contact angle results indicate that the hydrophilic layer favors OER, while the hydrophobic layer favors ORR. The synergies between these two layers enable rapid oxygen desorption during OER. They accelerate oxygen consumption from the air during ORR within the RuSA-NiFe LDH HE, overcoming the challenge posed by the conflicting three-phase interfaces in the air cathode. Optical microscopy images illustrate the formation of numerous tiny oxygen bubbles on the RuSA-NiFe LDH HE at various current densities (Fig. 4i and Supplementary Fig. 35), while the traditional RuSA-NiFe LDH air cathode exhibits large bubbles (Fig. 4j). At the current density of 50 mA cm−2, the average diameter of oxygen bubbles generated on the RuSA-NiFe LDH HE is mainly 15 μm (Supplementary Fig. 35c), smaller than that observed on the RuSA-NiFe LDH air cathode (25 μm, Supplementary Fig. 35d). In-situ optical microscopy video reveals that the timely release of tiny oxygen bubbles coming out from the RuSA-NiFe LDH HE surfaces even at high current densities (Supplementary Fig. 36 and Movie 1). In contrast, the traditional RuSA-NiFe LDH air cathode releases large oxygen bubbles at a slower pace. The prompt removal of tiny oxygen bubbles from the air cathode surface prevents the formation of dead areas, enhancing oxygen transfer efficiency.

The oxygen bubbles would cause uneven electric field distribution, which leads to nonuniform Zn deposition. The morphology of the Zn plate after 300 h was examined via a laser confocal scanning microscope (LCSM). The 3D LCSM images demonstrate that the pristine Zn plate shows a smooth surface (Supplementary Fig. 37). When RuSA-NiFe LDH HE is applied as the air cathode, the Zn plate preserves a relatively smooth morphology with a maximum height of 360 μm after 300 h cycling (Fig. 4k). Conversely, when the traditional RuSA-NiFe LDH air cathode is used, the Zn plate displays severe Zn dendrites, with island-like Zn dendrites reaching a maximum height of 2180 μm (Fig. 4l). To elucidate the impact of oxygen bubble behavior on Zn anode evolution, finite element simulations were conducted concerning oxygen bubble size and electric field distribution. The presence of a hydrophilic layer facilitates the rapid nucleation and subsequent desorption of minute oxygen bubbles, as evidenced in Fig. 4m. Conversely, oxygen bubbles merge to form a large bubble and desorb slowly from the traditional hydrophobic air cathode (Fig. 4n). The generated oxygen bubbles also induce significant electric field turbulence, resulting in irregular electric field distribution and uneven Zn deposition. Consequently, when paired with the air cathode with the hydrophilic layer, the Zn anode evolves to a slightly rough surface during extended cycles in the simulated model. In contrast, the traditional air cathode without the hydrophilic layer leads to substantial Zn dendrites on the side opposite the oxygen bubble edge. We further delved into and quantified the mechanism by which bubble evolution impacts Zn deposition (Fig. 4o, p), simulating the evolution of local current density distribution on the anode following several cycles of bubble generation. The closer the bubbles are to the Zn anode, the greater the perturbation on Zn deposition. When employing conventional hydrophobic air cathodes that engender oversized bubbles, their sluggish desorption behaviors lead to pronounced variations in local current density during impinging upon the anode interfaces. Conversely, incorporating RuSA-NiFe LDH HE significantly accelerates oxygen bubble detachment, ensuring bubbles vanish before reaching the anode interfaces. This timely desorption effectively diminishes considerable adverse effects on the local current density distribution at the Zn anode.

To definite the role of the chameleon-like bifunctional catalyst in the air cathode, we developed some characterization techniques to study the oxygen bubble diffusion, carbon corrosion and voltage fluctuation at the air cathode (Fig. 5a). Using gas chromatography, we developed a drainage system to collect the desorption oxygen during charge and conducted a quantitative analysis of the oxygen released from the cathode. The RuSA-NiFe LDH HE releases an oxygen content of 84.7% compared to the theoretical value (Supplementary Table 4), surpassing that of traditional RuSA-NiFe LDH air cathode (57.3%) at a current density of 10 mA cm−2 over an hour (Fig. 5b). This underscores the role of the hydrophilic layer in accelerating oxygen desorption. The pristine electrolyte is colorless, but it turns to a dark yellow electrolyte under high charge polarization voltage (Fig. 5c inset), which indicates the presence of side-reaction. To address the energy efficiency concerns in the air cathode, we tracked the evolution of carbon content in the electrolyte. The inorganic carbon relative ratio increases to 325%, 633%, and 2110% when utilizing RuSA-NiFe LDH HE, traditional RuSA-NiFe LDH air cathode, and carbon black air cathode (Fig. 5c), respectively. However, the organic carbon content in the electrolyte remains relatively stable. These findings indicate that (1) the carbon in the cathode is oxidized to inorganic carbon species but not organic carbon; (2) higher polarization voltage generates more inorganic carbon, which leads to battery degradation.

a Schematic illustration of techniques to reveal the oxygen bubbles, carbon corrosion and potential fluctuation on the air cathode. b The released to theoretical oxygen ratio of RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode were measured by gas chromatography. c The inorganic and organic carbon contents of pristine electrolytes and the electrolytes with different air cathodes after cycling at 10 mA cm−2 for 200 h, the high purity oxygen was used as the oxygen source. d The corrosion depth distribution evolution at the cathode surface upon charging. e The concentration of corrosion product evolution in the electrolyte during the charging process. f, g SEM images of the hydrophobic layer in RuSA-NiFe LDH HE (f) and traditional RuSA-NiFe LDH air cathode (g) after cycled in the batteries for 2000 h. For the SEM test, the hydrophobic layer and hydrophilic layer are close together instead of rolling together. h The fluctuation of cathode and anode potentials during discharge and cathode with a Hg/HgO electrode as the reference electrode. i Comparison of cathode potentials of RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode at various charge current densities. Inset is the cathode charge potential differences between RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode.

Moreover, we also employed finite element simulations to elucidate the mechanism of organic carbon formation within the electrolyte under high charge voltage with a current density of 100 mA cm−2. The high OER/ORR performance of RuSA-NiFe LDH HE significantly diminishes battery polarization, thus preventing the attainment of the cathode corrosion equilibrium potential that would otherwise induce corrosion (Fig. 5d). In contrast, the conventional RuSA-NiFe LDH air cathode exhibits severe morphology corrosion due to elevated polarization levels that trigger detrimental side reactions. Concurrently, this leads to greater corrosion byproducts (inorganic carbon) entering the electrolyte (Fig. 5e), contaminating subsequent reactions. Additionally, SEM images show that the pristine air cathode has a smooth surface (Supplementary Fig. 38), and the RuSA-NiFe LDH HE (Fig. 5f) kept a smooth surface after cycling for 2000 h. However, some cracks are observed in the conventional RuSA-NiFe LDH air cathode (Fig. 5g) after cycled for 2000 h. Therefore, simulation and experiment results reveal that RuSA-NiFe LDH HE effectively suppresses cathode corrosion and restricts the ingress of corrosion products into the electrolyte, consistent with the experimental results.

Monitoring cathode and anode potentials during battery operation is vital for understanding the significant charge/discharge voltage gap in RZABs and offering guidelines to narrow the gap. We incorporated a Hg/HgO reference electrode into the potential-monitoring cell to in-situ track cathode and anode potentials during both charge and discharge processes. The Hg/HgO electrode was chosen as a reference due to its stable potential in an alkaline solution. We observe that the cathode potential is −0.207 V (vs. Hg/HgO) during discharge and 0.484 V during charge (Fig. 5h), while the anode potentials are −1.372 V and −1.393 V during discharge and charge, respectively. The significant difference in cathode potential during discharge and charge suggests that the primary fluctuation in battery voltage stems from OER/ORR polarization. During discharge, the RZABs with RuSA-NiFe LDH HE show almost the same cathode potential as those with traditional RuSA-NiFe LDH air cathode with increasing current density (Supplementary Fig. 39). Upon switching to charge, RZABs with RuSA-NiFe LDH HE display a lower cathode potential compared to those with traditional RuSA-NiFe LDH air cathode. Even at a high current density of 100 mA cm−2, RuSA-NiFe LDH HE consistently delivers a stable charge potential, while the conventional RuSA-NiFe LDH air cathode experiences fluctuating potential due to sluggish oxygen desorption. The voltage difference increases linearly with the rise in current density (Fig. 5i), highlighting the role of the hydrophilic layer in improving OER kinetics. During charge, RZABs with RuSA-NiFe LDH HE demonstrate a higher anode potential compared to those with traditional RuSA-NiFe LDH air cathode, indicating a low overpotential that suppresses Zn dendrite growth. This result also reveals that the hydrophilic layer enables fast oxygen desorption and prevents severe polarization. This operando potential-monitoring system offers crucial insights into unraveling battery polarization mechanisms.

Despite the alluringly high theoretical specific energy of RZABs, it is becoming increasingly evident that achieving substantial charge/discharge capacity per cycle necessitates enhancing the depth of discharge (DOD) at lean electrolytes. To illustrate the potential applications of RuSA-NiFe LDH HE, we assembled RZABs utilizing RuSA-NiFe LDH HE and compared them with other air cathodes. The RuSA-NiFe LDH HE shows a notably high open circuit voltage (OCV) of 1.573 V (Fig. 6a and Supplementary Fig. 40), surpassing the values observed for RuSA-NiFe LDH (1.49 V), Pt/C + IrO2/C (1.38 V), and NiFe LDH (1.33 V). RuSA-NiFe LDH HE displays a maximum power density of 299.2 mW cm−2 under oxygen flow (Fig. 6b) and outperforms RuSA-NiFe LDH (265.1 mW cm−2) and Pt/C + IrO2/C (157.7 mW cm−2). Additionally, when utilizing air as the oxygen source, RuSA-NiFe LDH HE exhibits a higher powder density compared to RuSA-NiFe LDH (Supplementary Fig. 41a). This underscores that the hydrophilic layer improves oxygen utilization and facilitates the hydroxide ion diffusion. The RuSA-NiFe LDH HE showcases a remarkable specific capacity of 803.2 mAh g−1Zn, corresponding to an specific energy of 963.7 Wh kg−1Zn, superior to RuSA-NiFe LDH and Pt/C + IrO2/C (Supplementary Fig. 41b). The suitable OER interface provides by the hydrophilic layer results in gentle voltage increases with increasing charge current density, leading to lower charge voltages compared to RuSA-NiFe LDH and Pt/C + IrO2/C at any given current densities (Fig. 6b). The fluctuations observed in the charge polarization curves are primarily due to the oxidation of NiFe LDH to NiFeOOH. Even at a high charge current density of 300 mA cm−2, RuSA-NiFe LDH HE has a low voltage of 2.17 V, indicating its rapid charge capability. As the current density increases, RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode display similar discharge voltage drops (Fig. 6c). Meanwhile, the charge voltage for RuSA-NiFe LDH HE slightly increases. In contrast, the RuSA-NiFe LDH air cathode significantly increases, underscoring the hydrophilic layer favors charging. Given that the energy efficiency issue in RZABs stems from a large charge/discharge voltage gap, RuSA-NiFe LDH HE narrows this gap to achieve high energy efficiency. The RuSA-NiFe LDH HE demonstrates high energy efficiency of 68.86%, 61.2%, 55.1%, 45.4%, and 68.92% at 10, 25, 50, 100, and 10 mA cm−2, respectively. Operating temperature tolerance is a crucial factor in evaluating the practicality of RZABs. Decreasing temperature increases electrolyte viscosity, slowing down oxygen desorption and ion diffusion. RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode exhibit an increasing charge/discharge voltage gap at low temperatures. Notably, RuSA-NiFe LDH HE displays a smaller charge voltage difference compared to the traditional RuSA-NiFe LDH air cathode as the temperature dropped from 10 °C to −35 °C (Fig. 6d), emphasizing the hydrophilic layer’s rapid oxygen desorption at low temperatures. Subsequently, as the temperature rises to 60 °C, the charge voltage gap between RuSA-NiFe LDH HE and the traditional RuSA-NiFe LDH air cathode diminishes. These findings demonstrate that the hydrophilic layer accelerates oxygen desorption, significantly reducing charge voltage even under high current density and low-temperature conditions. The discharge/charge capacity per cycle and cycle life are pivotal factors in assessing battery practicability. RZABs with RuSA-NiFe LDH HE (Fig. 6e) exhibit a long cycle life of 2400 h without noticeable performance degradation at a high discharge/charge capacity of 100 mAh cm−2 per cycle. The SEM images (Supplementary Figs. 42 and 43) show that both the RuSA-NiFe LDH/CF layer and the hydrophobic layer remained well-preserved after extended cycling in RZABs, demonstrating the high structural stability of RuSA-NiFe LDH HE. Additionally, the HAADF-STEM, EDS and XPS results (Supplementary Figs. 44 and 45) confirm that the RuSA-NiFe LDH retained its single-atom Ru structure and primary Ni and Fe chemical states, further highlighting the high stability of RuSA-NiFe LDH. Ah-level zinc-air pouch cell with RuSA-NiFe LDH HE delivers a cell capacity of 1.3 Ah with a high DOD of 93.1% (Fig. 6f), The depth of discharge is calculated based on the consumption of Zn relative to the amount of pristine Zn powder. The specific energy of zinc-air pouch cell is 271 Wh kg−1 based on mass of active materials, anode current collectors, separator, air cathode and electrolyte (Supplementary Table 5), which outperforms most of the reported aqueous batteries. The RZABs with RuSA-NiFe LDH HE show a high cumulative area capacity (Fig. 6g) and good comprehensive performance (Fig. 6h), superior to most reported RZABs with other air cathodes (Supplementary Tables 6 and 7). To showcase the potential application of RZABs with RuSA-NiFe LDH HE for renewable energy storage systems, we assembled a scaled-up RZAB pack (Supplementary Fig. 46) for storing solar energy. As proof of concept, we assembled a system with an RZAB pack (24 V, 40 Ah) and a solar panel through an inverter. This integrated system stored solar energy during the day and powered electric appliances around the clock. The system provides energy for cooking during the day and a barbecue at night without any external electrical input (Supplementary Fig. 47 and Movie 2).

a OCV curves of RZABs based on various cathodes. The inset shows that a multimeter tested the OCV of Zn||RuSA-NiFe LDH HE. b The polarization curves and corresponding power density plots of RZABs with RuSA-NiFe LDH HE, traditional RuSA-NiFe LDH air cathode, and Pt/C + IrO2/C air cathode. c Galvanostatic charge/discharge curves of RZABs with RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode at 10, 25, 50, 100 mA cm−2 and return to 10 mA cm−2, the charge and discharge time is 30 min each cycle. d Galvanostatic charge/discharge cycling performance of RZABs with RuSA-NiFe LDH HE and traditional RuSA-NiFe LDH air cathode at 10 mA cm−2 and the temperature range of −35 °C to 60 °C. e Long-term cycling curves of RZABs with RuSA-NiFe LDH HE at 10 mA cm−2 with a high discharge/charge capacity of 100 mAh cm−2. f Discharge performance of Zn-air pouch cell with RuSA-NiFe LDH HE, insert optical image of ampere-hour Zn-air pouch cell, the specific energy is given based on the mass of active materials, Ti mesh, separator, air cathode and electrolyte. g Area capacity per cycle versus cumulative capacity of our RZAB with RuSA-NiFe LDH HE and other reported catalysts. h Comparison of the critical performances of RZAB with RuSA-NiFe LDH HE and other RZABs.

In summary, we introduce a chameleon-like bifunctional catalyst capable of self-adaptively adjusting active sites for ORR and OER, facilitating reversible transformation between RuSA-NiFe LDH and RuSA-NiFeOOH within RZABs. The RuSA-NiFe LDH is an ideal bifunctional catalyst, adept at self-adjusting active sites for oxygen reversible reactions, while demonstrating high catalytic activity and stability. It has been shown to optimize discharge/charge capacity per cycle and improve rate performance by a hierarchical air cathode with an additional hydrophilic layer atop the conventional air cathode. The RuSA-NiFe LDH HE with optimal gas-liquid-solid three-phase interfaces facilitated the reversible conversion between RuSA-NiFe LDH and RuSA-NiFeOOH, accelerated oxygen desorption, and lowered charge polarization voltage. The high discharge/charge capacity of 100 mAh cm−2 per cycle and a long lifespan of 2400 h were realized. An ampere-hour level zinc-air pouch cell achieves an impressive specific energy of 271 Wh kg−1, calculated based on the total mass of pouch cell, excluding the mass of packaging material. Furthermore, our large-scale RZAB pack demonstration underscores the practicality of RuSA-NiFe LDH HE for renewable energy storage applications. We have also developed various techniques to dissect the role and failure mechanisms of the air cathode. Overall, the chameleon-like bifunctional catalyst, hierarchical electrode design, and mechanism analysis techniques hold promise for enhancing performance in other batteries involving conversion and gas-involved reactions.

2.25 mg ruthenium chloride hydrate (RuCl3·xH2O, Aladdin, 95%), 243.6 mg nickel nitrate hexahydrate (Ni(NO3)2·6H2O, Aladdin, 98%), 101 mg Fe(NO3)3·9H2O (Aladdin, analytic reagent), 94.35 mg ammonium fluoride (NH4F, Aladdin, 96%) and 900 mg urea (CO(NH2)2, Aladdin, 99%) were homogenously dissolved in 36 mL deionized (DI) water. The as-obtained solution was transferred to a 50 mL Teflon-lined stainless-steel autoclave and kept at 160 °C for 5 h. The powder was washed with DI water for three-times and lyophilized for 48 h. The as-prepared sample was denoted as RuSA-NiFe LDH. For RuSA-NiFe LDH grown on the carbon felt (RuSA-NiFe LDH/CF), a piece of carbon felt (2 cm × 4 cm, Chener energy Co., Ltd.) was added, and the following processes are identical to the above for RuSA-NiFe LDH powder. For large-scale preparation of RuSA-NiFe LDH, the precursor dosage is amplified 2000 times and reacts at 100 L autoclave. The following processes are identical to those used for the above RuSA-NiFe LDH. For large-scale preparation of RuSA-NiFe LDH/CF, the same procedure was used with a roll of carbon felt (140 cm × 50 cm) as the substrate, and the hydrothermal reaction took place in a 50 L Teflon-lined autoclave. All aqueous solutions were prepared using deionized (DI) water with 18.2 MΩ cm resistance.

For comparison, RuNP-NiFe LDH and RuNP-NiFe LDH/CF were synthesized by the same procedure used for RuSA-NiFe LDH and RuSA-NiFe LDH/CF with the addition of 10 mg RuCl3·xH2O, respectively.

For comparison, NiFe LDH was synthesized by the same procedure used for RuSA-NiFe LDH without the addition of RuCl3·xH2O.

For RuSA-NiFe LDH electrode, 30 mg RuSA-NiFe LDH powder, 120 mg conductive carbon (VXC-72R, CABOT), and 450 μL polytetrafluoroethylene solution (PTFE, D-210C, DAIKIN, 5 wt% in water) were uniformly mixed in 5 ml ethanol in a mortar to form a paste, and rolled onto a carbon film (PLM01, Sike Co., Ltd.) with Ni mesh (95%, 60 mesh, Guangjiayuan Co., Ltd.) as a current collector. For the NiFe LDH electrode, the process was similar as that of the RuSA-NiFe LDH electrode. For the Pt/C + IrO2/C electrode, 15 mg Pt/C (20 wt%, Johnson Matthey), 15 mg IrO2/C (20 wt%, Johnson Matthey), 120 mg conductive carbon, and 450 μL 5 wt% PTFE were uniformly mixed in 5 mL ethanol and rolled onto a carbon film with Ni mesh as the current collector. This as-prepared electrode was then dried at 200 °C for 12 h under vacuum. The area mass loading of active materials was ~5 mg cm−2. For the carbon black electrode, the electrode was prepared without the additive of catalyst.

For RuSA-NiFe LDH hierarchical electrode (RuSA-NiFe LDH HE), a piece of RuSA-NiFe LDH/CF was rolled on the RuSA-NiFe LDH electrode and dried at 200 °C for 12 h under vacuum. The area loading mass of the catalyst is 15–22 mg cm−2 in the hierarchical cathode.

For the electrolyte containing 6 mol L−1 potassium hydroxide (KOH), 0.2 mol L−1 zinc acetate dihydrate (Zn(Ac)2·2H2O) and saturated zinc oxide (ZnO), 336.7 g of KOH pellets (Macklin, 95%) were slowly added to approximately 500 mL of deionized water while stirring continuously to ensure complete dissolution. Since the dissolution of KOH is highly exothermic, the solution was cooled in an ice bath. Afterward, 44.3 g of Zn(Ac)2·2H2O powder (Aladdin, 99%) was added to the alkaline solution and dissolved. Then, 25 g of ZnO powder (Aladdin, 99%) was added and stirred for 5 h. The cooled solution was transferred to a 1-liter volumetric flask, and deionized water was added to adjust the final volume to 1 L. The prepared electrolyte was thoroughly mixed before use and stored in a sealed container at room temperature (24–26 °C). For the 1 mol L−1 KOH electrolyte, 56 g of KOH pellets were slowly added to approximately 800 mL of deionized water while stirring continuously to ensure complete dissolution. The cooled solution was transferred to a 1-liter volumetric flask, and deionized water was added to adjust the final volume to 1 L. The electrolyte was thoroughly mixed before use and stored in a sealed container at room temperature (24–26 °C).

Field-emission scanning electron microscopy (SEM) images and corresponding energy-dispersive X-ray spectra (EDS) were obtained using a TESCAM MIRA3 system. The crystal structure was analyzed using the powder X-ray diffraction (XRD, Bruker, D8 advance) with Cu-Kα radiation (λ = 1.54 Å). Transmission electron microscopy (TEM) images were captured on an FEI Tecnai F30 (300 kV), while atomic resolution scanning TEM (STEM) images and corresponding EDS elemental mappings were acquired with a double Cs-corrector FEI Titan Themis G2 60-300 (300 kV). X-ray photoelectron spectroscopy (XPS) spectra were recorded with a PHI 5000 VersaProbe III (Al-Kα, hv = 1486.6 eV), and fitted using XPS Peak Fit (4.1) software. Raman spectra were collected with a LabRAM HR Evolution (HORIBA, 532 nm). The mass content of metals (Ru, Ni, and Fe) was measured via inductively coupled plasma optical emission spectrometry (ICP-OES, Agilent 7300). Electrolyte and O2 bubble contact angles were measured using a KRUSS DSA25 contact angle meter, the O2 bubbles released from a syringe needle. The contents of organic carbon and inorganic carbon in electrolytes were analyzed using a total organic carbon analyzer (enviro TOC) after dilution to near-neutral pH. Optical microscope images were recorded with an Mshot (MJ33). The surface area was evaluated by the Brunauer-Emmett-Teller (BET) method with a Micrometric ASAP 2020 system. Oxygen release from the cathode at a charge current density of 10 mA cm−2 for 1 h was analyzed by gas chromatography (GC-7920, ZhongJiaoJinYuan). The 3D morphology of the zinc anode was captured using a confocal laser scanning microscope (CLSM, VK-X1000, KEYENCE).

X-ray absorption spectra (XAS) were recorded on the BL01C1 beamline at NSRRC, with technical support provided by the Ceshigo Research Service Agency (www.ceshigo.com). The resulting XAFS data was processed using Athena software (version 0.9.26) for background correction, as well as pre-edge and post-edge line calibrations. Then, Fourier’s transformed fitting was carried out in Artemis (version 0.9.26). The k3 weighting, k-range of 3–~13.3 Å−1 and R range of 1–3 Å were used for the fitting of Ru foil; The k3 weighting, k-range of 3–~10 Å−1 and R range of 1–2 Å were used for the fitting of RuSA-NiFe LDH. The four parameters, such as coordination number (CN), bond length (R), Debye-Waller factor (σ2), and E0 shift (ΔE0), were fitted without anyone was fixed.

The electrochemical measurements were evaluated by a CHI 760E or Solartron 1260/1287A electrochemical workstation at room temperature (24–26 °C). A standard Hg/HgO electrode (R0501, Tianjin AIDA Co., Ltd.) and a graphite rod electrode (C305, Tianjin AIDA Co., Ltd.) were used as the reference and counter electrodes, respectively. The Hg/HgO reference electrode was calibrated for reversible hydrogen electrode (RHE) with a Pt wire in a high-purity H2-saturated electrolyte. The potentials of oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) were calibrated versus RHE. All ORR and OER polarization curves were given with iR compensation. Electrochemical impedance spectroscopy (EIS) measurements were made using an AC voltage with an amplitude of 5 mV in the frequency range of 100 kHz to 0.1 Hz. The resistance of the electrolyte was obtained from EIS results. The current density was given based on the geometric area of the electrode. Polarization curves were obtained by linear sweep voltammetry (LSV) with a scan rate of 5 mV s−1. The OER and ORR electrochemical activities were tested in 1 mol L−1 KOH (pH = 14 ± 0.2) and O2-saturated 0.1 mol L−1 KOH (pH = 13 ± 0.2), respectively. For static and rotating three-electrode tests, the type of electrochemical cell are C001 (100 mL, Tianjin AIDA Co., Ltd.) and C011 (150 mL, Tianjin AIDA Co., Ltd.), respectively. Each electrochemical test was performed in triplicate, and the presented graphical data were derived from the mean values of these replicates, with deviations of less than 1%.

For the free-standing electrode, the electrode was cut to 1 cm × 1 cm and directly used as the working electrode for the electrochemical tests. For comparison, a homogeneous catalyst ink with 5 mg IrO2/C, 460 μL isopropanol (Aladdin, 95%), 460 μL ethanol (Aladdin, 95%), and 80 μL Nafion solution (D520, DuPont, 5 wt% in water) was prepared by sonicating for 2 h. The IrO2/C was loaded on the surface of a carbon felt (1 cm × 1 cm) with a loaded mass of ~ 6 mg cm−2. For the powder catalyst, 5 mg catalyst was dispersed in 950 μL ethanol/isopropanol (1:1 v/v) and 50 μL Nafion (5 wt%) by ultrasonication for 4 h to obtain a homogeneous ink. The ink was dropped on a glassy carbon electrode and dried in the air. The mass loading of the catalyst was 0.255 mg cm−2. The working electrodes were activated using cycle voltammetry (CV) tests at a scan rate of 50 mV s−1 several times until the CV curves were stable. To eliminate the effects of oxidation peaks and accurately estimate the OER activity, the LSV curves from the positive to negative potential were used to analyze the overpotential. For the durability test, the potentials were collected at various current densities.

A homogeneous catalyst ink was prepared by dispersing 5 mg of catalyst in 475 μL ethanol, 475 μL isopropanol, and 50 μL of 5 wt% Nafion solution, followed by 4 h of ultrasonication. The ink was then drop-cast onto a glassy carbon electrode (5 mm diameter) and air-dried, achieving a catalyst loading of 0.255 mg cm−2. Electrochemical tests of the ORR catalysts were carried out using a rotating disk electrode (RDE, PINE) at various rotation speeds. For durability assessment, the i–t curve was recorded at a fixed potential under an O2 flow.

The kinetic current density (jk, mA cm−2) was calculated using the Koutecky–Levich Eq. (1).

where j is the measured current density (mA cm−2), and jL is the diffusion-limiting current density (mA cm−2).

The rotating ring-disk electrode (RRDE, E7R9, PINE) has a GC disk (Φ 5.61 mm) and Pt ring (inner Φ 6.25 mm and outer Φ 7.92 mm). For the RRDE tests, the curves were recorded in O2-saturated 0.1 mol L−1 KOH, and the ring potential was set to 1.5 V (vs. RHE) with a rotation speed of 1600 rpm. The electron transfer number (n) and yield of hydrogen peroxide ion (HO2−%) were determined by Eqs. (2) and (3), respectively.

where Id and Ir are the disk current and ring current, respectively. N is the current collection efficiency of the Pt ring (N = 0.37).

The performance and stability measurements of the aqueous zinc-air battery were evaluated in home-made RZABs using a Neware battery test system (CT-4008T) at room temperature (24–26 °C), unless otherwise specified the different temperature. The relative humidity is not controlled and fluctuates with atmospheric conditions, ranging from 30% to 90%. The RZAB was assembled using a Zn plate (thickness 0.5 mm, 2.5 cm × 2.5 cm) as the anode and an air cathode as the cathode (1 cm × 1 cm). The cycle performance of RZABs was tested in the air unless it is mentioned that the RZABs were tested under oxygen flow. An aqueous solution containing 6 mol L−1 KOH, 0.2 mol L−1 Zn(Ac)2·2H2O and saturated ZnO was used as the electrolyte (pH = 14.5 ± 0.3, indicating that it is greater than 14, so the exact pH value may not be very precise). 9 mL electrolyte was used for RZABs, and the electrolyte would not circulate. The zinc plate and electrolyte both are replaced with a new one after the RZABs cycled for around 300–400 h. The current density was given based on the geometric area of the cathode. No separator was used in the home-made RZABs. For the home-made RZABs, the area loading mass of the catalyst are approximately 5 and 15 mg cm−2 in the conventional and hierarchical cathodes, respectively. Three independent battery tests were conducted to derive the average values and evaluate battery performance.

Zinc-air pouch cells were constructed with two RuSA-NiFe LDH HE (2 × 4 cm2) air cathodes and one Zn paste anode (2 × 4 cm2). Ni and Ti mesh were used as current collectors for the cathode and anode, respectively. An aqueous solution with 6 mol L−1 KOH, 0.2 mol L−1 Zn(Ac)2·2H2O and saturated ZnO was used as the electrolyte. The Al plastic film was used as the package material. For the Zn paste anode, 90 g of Zn powder (Zhongkeyannuo Co., Ltd., 99.9 %), 5 g of ZnO, 1 g of reolosil (SiO2, 7–40 nm, Aladdin, 99.8%) and 80 g of PTFE (5 wt%) were added to 50 mL of DI water and vigorously stirred to form a homogeneous slurry. The slurry was then cast onto the Ti mesh current collector and dried overnight at 65 °C in a vacuum oven. The nonwoven separator (FPC3018, 10 × 4.3 cm2, Laizhou lianyou jinhao new material Co., LTD) was used to separate the anode and cathode. For the pouch cell, the area loading mass of the catalyst is approximately 15 mg cm−2 in the hierarchical cathode.

Large-scale RZABs were assembled with two pieces of hierarchical electrode (250 cm2 per piece) on two sides and a zinc anode (thickness 5 mm, 250 cm2) in the middle. A nonwoven separator (FPC3018, 500 cm2) was placed between the anode and the air cathode. Ti and Ni wires were used as anode and cathode taps, respectively. To obtain an RZAB pack (24 V, 40 Ah), 20 RZABs were connected in series and then in parallel with another serial RZABs pack. An aqueous solution containing 6 mol L−1 KOH, 0.2 mol L−1 Zn(Ac)2·2H2O and saturated ZnO was used as the electrolyte. 10 L electrolyte was added into the container, the electrolyte is flowing during the battery work. The RZAB pack integrates with a monocrystalline silicon solar panel (48 V, 300 W, Anhui hongyi energy Co., LTD.), and stored the energy from solar panel. No iR correction was applied during any of the zinc-air battery tests. For the large-scale RZABs, the area loading mass of the catalyst is approximately 22 mg cm−2 in the hierarchical cathode.

The in-situ electrochemical reaction was performed with a CHI 760E electrochemical workstation and was imaged using a light microscope (MJ33).

To eliminate the signal interference from carbon, the RuSA-NiFe LDH electrode was prepared without conductive carbon. A silicon wafer was used to calibrate the spectral shift. The in-situ Raman spectra were collected while the cell was cycling at a galvanostatic charge/discharge current densities of 10 mA cm−2. To obtain the spectrum with a high ratio of signal to noise, the acquisition time of every spectrum was 3 min.

The electrode was prepared the same as the electrode for the in-situ Raman test. The electrode was assembled in RZABs and cycled for 200 h. Then, the electrode was obtained after the charge or discharge process.

The Hg/HgO reference electrode was calibrated for RHE with a Pt wire in a high purity H2-saturated 6 mol L−1 KOH solution. The cathode potential was recorded versus the potential of the Hg/HgO reference electrode. The anode potential was obtained by Eq. (4).

Where \({\varphi }_{{\rm{anode}}}\) is the anode potential, \({\varphi }_{{\rm{cathode}}}\) is the cathode potential, E is the battery voltage.

The total organic carbon analyzer directly measured the total carbon content and inorganic carbon content, while the organic carbon content was calculated using Eq. (5).

where COC, CTC and CIC represent the organic carbon, total carbon and inorganic carbon concentrations, respectively.

Based on the COMSOL Multiphysics 6.0 platform, the finite element analysis was divided into two major parts, the first of which pertains to bubble evolution, electric filed distribution and Zn deposition. The second part focuses on simulating cathode corrosion. The following are the main governing equations for simulation:

For the first part, the bubble evolution can be described using a volume of fluid (VF) with two-phase flow, the VF fractions (α, 0 ≤ α ≤ 1) representing the gas phase (VF1) and electrolyte phase (VF2) in this model. It denotes that the cell is full of the gas phase and electrolyte phase if the α = 1 and α = 0, respectively. Based on α in this model, the variable and tracking of the interfaces between the phases will be calculated in each control volume. Therefore, the phase field equation are the following Eqs. (6) and (7).

where ϕ is the phase variable, μ is the fluid flow velocity vector, εpf is the interface thickness control parameter, ψ is the intermediate variable in the phase field, λ is 3εpfσ/\(\sqrt{8}\), γ is a \({{{\rm{\chi }}}{\varepsilon }}_{{{\rm{pf}}}}^{2}\), χ is the migration mobility adjustment parameter, ∇φ is the gradient of the external free energy, σ is the surface tension, and t is time.

The Navier–Stokes Eqs. (8) and (9) describe the incompressible flow.

Where p is the pressure, \(\vec{{{\boldsymbol{\mu }}}}\) is the dynamic viscosity of the electrolyte, ρ is the density of the electrolyte, \(\vec{{{\bf{g}}}}\) is the gravitational acceleration, and \(\vec{{{{\boldsymbol{F}}}}_{{{\boldsymbol{st}}}}}\) is the surface tension.

The current conservation Eq. (10) describes the current in the fluid region.

Where σ is the material conductivity, and V is the electric potential.

The transport behavior of Zn2+ in the electrolyte can be described by the Nernst–Planck Eq. (11).

where NZn2+ is Zn2+ flux, DZn2+ is the diffusion coefficient, z is the number of transferred electrons, and \({c}_{0}\) is the concentration of Zn2+. The constants F and R are Faraday’s and ideal gas constant, respectively, and T is the absolute temperature in Kelvin, and Φ is the potential of the electrolyte.

Meanwhile, all ions within the electrolyte satisfy both charge conservation and mass conservation.

At the anode surface, the deposition of Zn can be simplified as Eq. (12).

The deposition rate of Zn can be quantitatively described and analyzed using the Buttler–Volmer Eq. (13).

where i0 is the exchange current density, η is the overpotential, αa, and αc are the anodic and cathodic charge transfer coefficients, and cZn2+ is the Zn2+ concentration near the anode surface.

For the phase field boundary, at the bottom of the liquid-solving domain, bubbles are generated and gradually rise under the influence of gravity. The sides are set as no-flux boundaries, and the top is set as an outlet boundary condition. For the flow boundary at the bottom of the liquid-solving domain it is designated as an inlet boundary condition, the sides are set as no-slip wall boundaries, and the top is set as an outlet boundary condition. The current boundary is set at the bottom as a current-in boundary condition, and at the top, it is designated as a ground boundary condition.

For the second part, simulating cathode corrosion, we established a different three-dimensional (3D) model to visualize the morphology of cathode corrosion. When the overpotential exceeds 1.85 V, side reactions occur at the cathode, leading to the formation of byproducts that enter into the electrolyte. The governing equation for these side reactions is also subject to the Butler-Volmer equation in Eq. (13). Moreover, to simulate the non-uniformity of corrosion, a randomly distributed function (Eq. (14)) is applied to the cathode boundary49.

where x and y represent the spatial coordinates, m and n denote the spatial frequencies; \(a(m,n)\) represents the amplitude; and \(\varphi (m,n)\) denotes the phase angle. The amplitude values are calculated randomly according to a Gaussian distribution function, whereas the phase angles and spatial frequencies are obtained from a uniform random distribution within a specified range50.

All first-principles density functional theory (DFT) calculations were executed using the Vienna ab initio simulation package51. The projector augmented wave approach was adopted to describe the electron-core interaction52. The Perdew-Burke-Ernzerhof functional within a generalized gradient approximation was applied to describe the exchange-correlation energies53. A 400 eV cut off energy for the plane-wave basis set was used. The self-consistent filed calculation was performed with an atomic force convergence below 0.03 eV Å⁻¹. Brillouin zone sampling during structural optimization and self-consistent calculations was carried out using a Monkhorst-Pack k-space sampling grids of 2\(\times\)2\(\times\)1. Van der Waals interactions within water adsorption systems were considered using DFT-D3 method of Grimme54. To consider the strongly correlated d-orbital of transition metal atoms, the DFT + U method was applied for Coulomb and exchange corrections. Specific energy-exchange energy (U-J) values were assigned to the 3 d electrons of V (2.7 eV), Cr (3.5 eV), Mn (4.0 eV), Co (5.0 eV), Cu (4.0 eV), Zn (7.5 eV), Mo (2.4 eV), Ru (3.0 eV), Rh (3.3 eV), Pd (3.6 eV), Ag (5.8 eV), Ce (5.5 eV), and Pt (4.3 eV) to capture d-electron effects on electronic properties. All atoms in the optimized structures exhibited negative formation energy without significant displacement.

Following the computational hydrogen electrode model, the OER and ORR activities can be determined by the adsorption-free energies of key intermediates. The elementary steps of the adsorbate evolution mechanism for OER and ORR are detailed below.

The OER process is typically summarized in four steps (Eqs. 15–18) in alkaline media.

The ORR process is usually summarized in four steps (Eqs. 19–22) in alkaline media.

where * represents the catalytic active site, and *OOH, *O, *OH denote the adsorbed intermediates.

To obtain the rate-limiting step of the OER and ORR, the Gibbs free energies of each elementary step are defined as the difference in free energy between the products and reactants, obtained from Eqs. (23) and (24).

where \({E}_{{ads}}\) is the adsorption energy of the intermediate on the catalyst. \({\Delta E}_{{ZPE}}\) is the difference in zero-point energy between the adsorption state and the gas state. \(T\) is the temperature (300 K). \(\Delta S\) is the difference in entropy between the adsorption state and the gas state. n is the number of electrons involved in each step, and U is the applied electrode potential.

The overpotentials for the OER and ORR reactions are obtained from Eqs. (25) and (26).

where \(\Delta {G}_{1},\,\Delta {G}_{2},\,\Delta {G}_{3},\) and \(\Delta {G}_{4}\) are free energies of Eqs. (15)–(18). The \(\Delta {G}_{5},\,\Delta {G}_{6},\,\Delta {G}_{7},\) and \(\Delta {G}_{8}\) are free energies of Eqs. (19)–(22).

For the volcano plots, the adsorption Gibbs free energy differences of *O and *OOH are the main rate-determining steps and are used as a descriptor for OER activity, and the adsorption Gibbs free energy of *OH is used as a descriptor for ORR activity. The \({\eta }_{{\rm{OER}}}\) is stated as Eq. (27), and \({\eta }_{{ORR}}\) is stated as Eqs. (28) and (29)

The molecular formulas for RuSA-NiFe LDH, RuSA-NiFeOOH, and RuO3-NiFeO2 are Ni24Fe8RuO65H61, Ni24Fe8RuO65H48, and Ni24Fe8RuO65H32, respectively. The corresponding conversion reactions are described by Equations (30) and (31).

The computational hydrogen electrode energy is calculated using Eq. (32).

The Gibbs free energy change of conversion reaction is given by Eq. (33).

where \(\Delta G\) is the different in Gibbs free energy between the reactant and product Gibbs free energy. \({G}_{R}\) and \({G}_{p}\) represent the Gibbs free energy of reactant and product, respectively. \(G\left({H}_{2}\right)\) is the Gibbs free energy of an H2 molecule in the gas phase, n is the number of electrons, and U is the applied electrode potential.

A negative ΔG indicates a spontaneous reaction, and by solving Eq. (33), we can calculate the relationship between potential U and pH, and thus construct the Pourbaix diagram.

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Source data are provided with this paper. Any additional requests for information can be directed to, and will be fulfilled by, the corresponding authors. Source data are provided with this paper.

Sun, W. et al. A rechargeable zinc-air battery based on zinc peroxide chemistry. Science 371, 46–51 (2021).

Article ADS CAS PubMed Google Scholar

Shinde, S. S. et al. Ampere-hour-scale zinc–air pouch cells. Nat. Energy 6, 592–604 (2021).

Article ADS CAS Google Scholar

Zhou, G., Chen, H. & Cui, Y. Formulating energy density for designing practical lithium–sulfur batteries. Nat. Energy 7, 312–319 (2022).

Article ADS CAS Google Scholar

Ye, L. et al. A rechargeable calcium–oxygen battery that operates at room temperature. Nature 626, 313–318 (2024).

Article ADS CAS PubMed Google Scholar

Wang, Q., Kaushik, S., Xiao, X. & Xu, Q. Sustainable zinc-air battery chemistry: advances, challenges and prospects. Chem. Soc. Rev 52, 6139–6190 (2023).

Article CAS PubMed Google Scholar

Zhong, X. et al. Flexible zinc-air batteries with ampere-hour capacities and wide-temperature adaptabilities. Adv. Mater. 35, 2209980 (2023).

Article CAS Google Scholar

Zhou, W. et al. Regulating the scaling relationship for high catalytic kinetics and selectivity of the oxygen reduction reaction. Nat. Commun. 13, 6414 (2022).

Article ADS CAS PubMed PubMed Central Google Scholar

Zheng, Z. et al. An extended substrate screening strategy enabling a low lattice mismatch for highly reversible zinc anodes. Nat. Commun. 15, 753 (2024).

He, Y. et al. Boosting gaseous oxygen transport in a Zn-air battery for high-rate charging by a bubble diode-inspired air electrode. Energy Storage Mater. 57, 360–370 (2023).

Article Google Scholar

Zhong, X. W. et al. Engineering Pt and Fe dual-metal single atoms anchored on nitrogen-doped carbon with high activity and durability towards oxygen reduction reaction for zinc-air battery. Appl. Catal. B 286, 119891 (2021).

Article CAS Google Scholar

Deng, Y. P. et al. Dynamic electrocatalyst with current-driven oxyhydroxide shell for rechargeable zinc-air battery. Nat. Commun. 11, 1952 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Zhang, P. et al. Bifunctional single atom catalysts for rechargeable zinc-air batteries: from dynamic mechanism to rational design. Adv. Mater. 35, e2303243 (2023).

Article PubMed Google Scholar

Wu, S. et al. Single Cu–N4 sites enable atomic Fe clusters with high-performance oxygen reduction reactions. Energy Environ. Sci. 16, 3576–3586 (2023).

Article CAS Google Scholar

Wang, Y. et al. Regulation of atomic Fe-spin state by crystal field and magnetic field for enhanced oxygen electrocatalysis in rechargeable zinc-air batteries. Ange. Chem. Int. Ed. 62, e202304229 (2023).

Article CAS Google Scholar

Zhou, S. et al. Visualizing interfacial collective reaction behaviour of Li–S batteries. Nature 621, 75–81 (2023).

Article ADS CAS PubMed Google Scholar

Xiao, X. et al. Rational design of flexible Zn-based batteries for wearable electronic devices. ACS Nano 17, 1764–1802 (2023).

Article CAS PubMed Google Scholar

Yang, H. et al. Intramolecular hydroxyl nucleophilic attack pathway by a polymeric water oxidation catalyst with single cobalt sites. Nat. Catal. 5, 414–429 (2022).

Article CAS Google Scholar

Zhuang, Z. et al. p-Block-metal bismuth-based electrocatalysts featuring tunable selectivity for high-performance oxygen reduction reaction. Joule 7, 1003–1015 (2023).

Article CAS Google Scholar

He, Z. et al. Activating lattice oxygen in NiFe-based (oxy)hydroxide for water electrolysis. Nat. Commun. 13, 2191 (2022).

Article ADS CAS PubMed PubMed Central Google Scholar

Zhong, X. W. et al. Rechargeable zinc-air batteries with an ultralarge discharge capacity per cycle and an ultralong cycle life. Adv. Mater. 35, 2301952 (2023).

Article CAS Google Scholar

Deng, Y. P. et al. Reconstructing 3d-metal electrocatalysts through anionic evolution in zinc-air batteries. J. Am. Chem. Soc. 145, 20248 (2023).

Article CAS PubMed Google Scholar

Cao, Q. et al. A fluorinated covalent organic framework with accelerated oxygen transfer nanochannels for high-performance zinc-air batteries. Adv. Mater. 35, e2210550 (2023).

Article PubMed Google Scholar

Chen, B., Zhong, X., Zhou, G., Zhao, N. & Cheng, H. M. Graphene‐supported atomically dispersed metals as bifunctional catalysts for next‐generation batteries based on conversion reactions. Adv. Mater. 34, 2105812 (2021).

Article Google Scholar

Lei, X. et al. High-entropy single-atom activated carbon catalysts for sustainable oxygen electrocatalysis. Nat. Sustain. 6, 816–826 (2023).

Article Google Scholar

Zhong, X. W. et al. Co single-atom anchored on Co3O4 and nitrogen-doped active carbon toward bifunctional catalyst for zinc-air batteries. Appl. Catal. B 260, 118188 (2020).

Article CAS Google Scholar

Sun, T. et al. Ferromagnetic single-atom spin catalyst for boosting water splitting. Nat. Nanotechnol. 18, 763–771 (2023).

Article ADS CAS PubMed Google Scholar

Lang, R. et al. Non defect-stabilized thermally stable single-atom catalyst. Nat. Commun. 10, 234 (2019).

Article ADS PubMed PubMed Central Google Scholar

Shi, Z. et al. Phase-dependent growth of Pt on MoS2 for highly efficient H2 evolution. Nature 621, 300–305 (2023).

Article ADS CAS PubMed Google Scholar

Gloag, L., Somerville, S. V., Gooding, J. J. & Tilley, R. D. Co-catalytic metal–support interactions in single-atom electrocatalysts. Nat. Rev. Mater. 9, 73–189 (2024).

Zheng, X. et al. Origin of enhanced water oxidation activity in an iridium single atom anchored on NiFe oxyhydroxide catalyst. Proc. Natl Acad. Sci. USA 118, e2101817118 (2021).

Article CAS PubMed PubMed Central Google Scholar

Zhang, T. et al. Pinpointing the axial ligand effect on platinum single-atom-catalyst towards efficient alkaline hydrogen evolution reaction. Nat. Commun. 13, 6875 (2022).

Article ADS CAS PubMed PubMed Central Google Scholar

Zhang, J. et al. Single-atom Au/NiFe layered double hydroxide electrocatalyst: probing the origin of activity for oxygen evolution reaction. J. Am. Chem. Soc. 140, 3876–3879 (2018).

Article CAS PubMed Google Scholar

Zhou, T. P., Zhang, N., Wu, C. Z. & Xie, Y. Surface/interface nanoengineering for rechargeable Zn-air batteries. Energy Environ. Sci. 13, 1132–1153 (2020).

Article CAS Google Scholar

Yan, L. et al. Engineering self‐supported hydrophobic–aerophilic air cathode with CoS/Fe3S4 nanoparticles embedded in S, N Co‐doped carbon plate arrays for long‐life rechargeable Zn–air batteries. Adv. Energy Mater. 13, 2204245 (2023).

Article CAS Google Scholar

Tang, K. et al. Hydrophobization engineering of the air-cathode catalyst for improved oxygen diffusion towards efficient zinc-air batteries. Ange. Chem. Int. Ed. 61, e202202671 (2022).

Article ADS CAS Google Scholar

Yu, J., Li, B. Q., Zhao, C. X., Liu, J. N. & Zhang, Q. Asymmetric air cathode design for enhanced interfacial electrocatalytic reactions in high-performance zinc-air batteries. Adv. Mater. 32, e1908488 (2020).

Article PubMed Google Scholar

Li, Y. et al. Advanced zinc-air batteries based on high-performance hybrid electrocatalysts. Nat. Commun. 4, 1805 (2013).

Article ADS PubMed Google Scholar

Cheng, W. et al. Lattice-strained metal–organic-framework arrays for bifunctional oxygen electrocatalysis. Nat. Energy 4, 115–122 (2019).

Article ADS CAS Google Scholar

Mehmood, A. et al. High loading of single atomic iron sites in Fe–NC oxygen reduction catalysts for proton exchange membrane fuel cells. Nat. Catal. 5, 311–323 (2022).

Article CAS Google Scholar

Yang, H. et al. Preparation of nickel-iron hydroxides by microorganism corrosion for efficient oxygen evolution. Nat. Commun. 11, 5075 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Liu, K. et al. Strong metal-support interaction promoted scalable production of thermally stable single-atom catalysts. Nat. Commun. 11, 1263 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Zhong, X. W. et al. 3D heterostructured pure and N-Doped Ni3S2/VS2 nanosheets for high efficient overall water splitting. Electrochimica Acta 269, 55–61 (2018).

Article CAS Google Scholar

Thorarinsdottir, A. E., Erdosy, D. P., Costentin, C., Mason, J. A. & Nocera, D. G. Enhanced activity for the oxygen reduction reaction in microporous water. Nat. Catal. 6, 425–434 (2023).

Article CAS Google Scholar

Zhao, C. X. et al. A clicking confinement strategy to fabricate transition metal single-atom sites for bifunctional oxygen electrocatalysis. Sci. Adv. 8, eabn5091 (2022).

Article MathSciNet CAS PubMed PubMed Central Google Scholar

Qiu, Z., Tai, C. W., Niklasson, G. A. & Edvinsson, T. Direct observation of active catalyst surface phases and the effect of dynamic self-optimization in NiFe-layered double hydroxides for alkaline water splitting. Energy Environ. Sci. 12, 572–581 (2019).

Article CAS Google Scholar

Lee, S., Banjac, K., Lingenfelder, M. & Hu, X. Oxygen isotope labeling experiments reveal different reaction sites for the oxygen evolution reaction on nickel and nickel iron oxides. Ange. Chem. Int. Ed. 58, 10295–10299 (2019).

Article CAS Google Scholar

Wu, Y., He, C. & Zhang, W. Capture-backdonation-recapture” mechanism for promoting N2 reduction by heteronuclear metal-free double-atom catalysts. J. Am. Chem. Soc. 144, 9344–9353 (2022).

Article CAS PubMed Google Scholar

He, C., Xu, C. & Zhang, W. Instructive synergistic effect of coordinating phosphorus in transition-metal-doped β-phosphorus carbide guiding the design of high-performance CO2RR electrocatalysts. ACS Appl. Mater. Inter. 15, 57015–57028 (2023).

CAS Google Scholar

Liu, Y. et al. Insight into the critical role of exchange current density on electrodeposition behavior of lithium metal. Adv. Sci. 8, 2003301 (2021).

Article CAS Google Scholar

Zhang, M. T., Qu, H. T. & Zhou, G. M. The factors that influence the electrochemical behavior of lithium metal anodes: electron transfer and Li-ion transport. N. Carbon Mater. 38, 776–783 (2023).

Article CAS Google Scholar

Kresse, G. & Furthmuller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).

Article ADS CAS Google Scholar

Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 59, 1758–1775 (1999).

Article ADS CAS Google Scholar

Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).

Article ADS CAS PubMed Google Scholar

Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).

Article ADS PubMed Google Scholar

Download references

This work was financially supported by the National Key Research and Development Program of China (Grant No. 2019YFA0705700), the National Natural Science Foundation of China (Grant Nos. 52072205 and 22309077), Guangdong Basic and Applied Basic Research Foundation (Grant Nos. 2023B1515120099 and 2022A1515110117), the Key Fundamental Research Project funding from Shenzhen Science and Technology Innovation Committee (Grants Nos. JCYJ20220818100406014 and JCYJ20200109141014474), and SUSTech Energy Institute for Carbon Neutrality (High level of special funds, G03034K001). The authors also thank Fengyi Zheng (SIGS), Zhexuan Liu (SIGS), Guangfu Luo (SUSTech) and Dongsheng He (SUSTech) for their discussion on this work and technical support. The authors would like to acknowledge the technical support from SUSTech Core Research Facilities and computing resources supported by the Center for Computational Science and Engineering at SUSTech.

These authors contributed equally: Xiongwei Zhong, Xiao Xiao.

Department of Materials Science and Engineering, and SUSTech Energy Institute for Carbon Neutrality, Southern University of Science and Technology, Shenzhen, 518055, China

Xiongwei Zhong, Zhitong Li, Leyi Gao, Xingzhu Wang & Baomin Xu

Shenzhen Geim Graphene Center, Tsinghua-Berkeley Shenzhen Institute & Shenzhen International Graduate School, Tsinghua University, Shenzhen, 518055, China

Xiao Xiao, Mengtian Zhang, Zhiyang Zheng, Qingjin Fu & Guangmin Zhou

National Graphene Institute, University of Manchester, Manchester, UK

Qizhen Li

School of Materials Science and Engineering, Tianjin University, Tianjin, 300350, China

Biao Chen

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

B.X., G.Z., and X.Z. conceived and designed this work. X.Z. conducted the experiments, and analyzed the results. Q.L. and X.W. conducted the TEM characterization and checked the data. M.Z. conducted the finite element simulation. X.X., L.G., Z.L., B.C., and Q.F. checked the data and calculation. X.Z., Z.Z., and X.X. prepared and organized the figures. X.Z., Q.L., X.X., Z.Z., G.Z., and B.X. wrote and revised the paper. G.Z. and B.X. served as technical leads for this work. All authors discussed the results and commented on the paper.

Correspondence to Xiongwei Zhong, Guangmin Zhou or Baomin Xu.

The authors declare no competing interests.

Nature Communications thanks Cheng He, Zhong-Yong Yuan, and the other anonymous reviewer(s)for their contribution to the peer review of this work. A peer review file is available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

Zhong, X., Xiao, X., Li, Q. et al. Understanding the active site in chameleon-like bifunctional catalyst for practical rechargeable zinc-air batteries. Nat Commun 15, 9616 (2024). https://doi.org/10.1038/s41467-024-54019-1

Download citation

Received: 26 April 2024

Accepted: 30 October 2024

Published: 07 November 2024

DOI: https://doi.org/10.1038/s41467-024-54019-1

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative