The ultrathin ZnS layer was galvanostatically deposited on the zinc surface in 0.2 M Na₂S solution. Upon the current flow (2.5 mA cm⁻²), the zinc metal underwent oxidation and released Zn²⁺, which then reacted with S²⁻ in the inner Helmholtz plane (IHP), forming the uniform ZnS SEI layer with a thickness of 200 nm over an electrodeposition time of 2 min (Figs. 1a and S1). It should be noted that the thickness is in the same order but lower than the theoretical value of 380 nm due to the lower current efficiency [34]. Importantly, the ZnS SEI layer demonstrates a strong adhesion to the Zn substrate, and even after being twisted, no cracks are observed on the ZnS surface (Fig. S2). The chemical composition of the ZnS film was analyzed by X-ray photoelectron spectroscopy (XPS). Figures 1b, c and S3 indicate the pristine Zn foil only presents the Zn 2p_1/2 and Zn 2p_3/2 signal with the binding energies of 1045 and 1022 eV, respectively [35]. In contrast, the Zn@ZnS electrode exhibits distinct S 2p_1/2 (162.4 eV) and S 2p_3/2 (161.3 eV) peaks, indicating the successful synthesis of ZnS [28]. Besides, the binding energy of zinc in the Zn@ZnS electrode shifts to a higher value due to the formation of Zn-S bonds. Good wettability of the electrolyte to the electrode is a crucial factor in optimizing the zinc nucleation behavior by advancing the Zn²⁺ diffusion toward the electrode and decreasing the nucleation energy (Eqs. S16 and S17) while retaining the nucleation radius [16,36]. Figure S4 shows the contact angle for the Zn@ZnS electrode (71°) is smaller, compared to the bare Zn electrode (106.5°), confirming a better wettability of the electrolyte toward Zn@ZnS electrode due to the stronger interaction with ZnS.

The morphology evolution before and after the ZnS SEI layer was studied by the scanning electron microscope (SEM) images. As depicted in Fig. 1d, e, the pure Zn surface displays numerous scratches and cracks due to its rough manufacturing. These imperfections induce the “tip effects” that unevenly distributed electric field aggravates the uniform deposition of Zn. With the ultrathin ZnS SEI layer (Fig. 1f, g), the electrode shows a flatter and smoother surface, and nanoparticle-like morphology can be observed. Such structure can increase the contact area of the electrode/electrolyte and guide the uniform diffusion and deposition of Zn. The atomic force microscope (AFM) images also suggest a smoother surface with the ZnS SEI layer where the surface average roughness (R_a) decreased from 24.3 nm for the bare Zn surface to 11.9 nm for ZnS covered surface (Fig. S5). In addition, the energy-dispersive X-ray spectroscopy (EDX) mapping images in Fig. 1h verified that S and Zn elements are homogeneously distributed across the surface, confirming the uniformity of the ZnS protection layer. The cross-sectional EDX mapping results shown in Fig. S6 also display the distribution of ZnS on the surface of the Zn electrode with an approximate thickness of 200 nm. The XRD results of the prepared Zn@ZnS electrode surface show no extra peaks except the Zn metal signal (Fig. S7). That may be due to the amorphous state of deposited ZnS. The thickness and morphology of this ZnS layer can be adjusted by controlling the electrodeposition current and time. Lower current density may cause the poor coating or high impurity of the deposits while higher current density tends to result in the severe agglomeration of deposits [37,38,39]. Besides, at a given capacity, a lower electrodeposition current generally corresponds to a thinner coating thickness due to the low current efficiency [37,38]. The SEM images in Fig. S8 exhibit that at a low current density of 1.25 mA cm⁻²_, cracks are still obvious on the surface, while with a high current density of 5 mA cm⁻², small nanoparticles appear on the surface. However, these particles are not uniform and tend to aggregate in some areas, which is consistent with the phenomenon observed in previous reports.

As stated by Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, the zinc ions in aqueous solutions suffer the electric repulsive force brought from the counterions in EDL as well as the van der Waals attractive force (VDW) [40]. Under the repulsive governed force, the zinc ions tend to deposit at “tips” through a 2D diffusion and form a loose structure, whereas a compact and dense structure can be obtained with attractive force. In this regard, a regulated EDL repulsion force is desired. According to the Poisson-Boltzmann (PB) equation, the reduced EDL repulsive force can be reflected in the suppressed EDL thickness, which can be described by Debye length [40].

To reveal the suppression of EDL by the presence of the ZnS SEI layer, multiple calculations were performed. Firstly, the EDL structure with and without the ZnS SEI layer was modeled by COMSOL Multiphysics based on the experimental data. As shown in Fig. 2a, the potential difference over the Helmholtz layer is higher (36.4 mV) for the Zn electrode than the Zn@ZnS electrode (22 mV), demonstrating that Zn²⁺ suffers a lower energy barrier for the desolvation and reduction in the presence of ZnS layer. This weakened EDL repulsive force is conducive to dense zinc deposition. The cell with Zn@ZnS electrode also shows a suppressed diffuse layer, indicating a reduced EDL repulsive force between the charged particles. The reduced EDL repulsive force can reduce the diffusion barrier for the hydrated zinc ions. Figure 2b, c intuitively displays the composition of EDL without and with the ZnS SEI layer. The hydrated zinc ions, along with SO₄²⁻ anions, can only approach the outer Helmholtz layer (OHL) [1,14]. Subsequently, the accompanied SO₄²⁻ anions, de-solvated Zn²⁺, and water molecules taken off from hydrates diffuse toward the zinc surface through the inner Helmholtz layer (IHL). With a bare Zn electrode, SO₄²⁻ anions and water molecules can specifically absorb on the surface and cover the active site for Zn²⁺ adsorption, leading to side reactions, and the formation of the by-product. Besides, under the high repulsion force, Zn²⁺ tends to diffuse and deposit at the protrusions due to the “tip effects”, forming scattered platelets and potentially inducing zinc dendrite growth (Fig. 2b). With ZnS film, the tips were covered and a uniform surface with a strong affinity to the Zn²⁺ can be obtained (Fig. 2c). The enhanced desolvation effects via the strong interaction between S and adatom could exclude water molecules from entering the IHP and prevent the direct contact of water and Zn substrate, forming a water-poor environment [14].

To experimentally study the charge state of zinc anode in an aqueous solution, the zeta potential was tested. The zeta potential of zinc powder in aqueous solutions exhibits a negative value, suggesting the zinc surface is negatively charged (Fig. 2d) [41,42]. By contrast, the zeta potential of ZnS powder is positive, with a high average value of 18.4 mV, consistent with the reported value [43]. The high zeta potential material is conducive to enhancing the electrokinetic effects of the battery by fastening the mass transport rate and adjusting ion distribution, which can be evidenced by the ions’ concentration gradient in the electrolyte (Fig. S9). Figure 2e shows the zeta potential of zinc deposits in pure Zn and Zn@ZnS surface. The results indicate both deposits present negative values, but a high absolute value of 12.2 mV for Zn@ZnS suggests the deposit is more stable from aggregating into a large nucleus, leading to uniform zinc deposition. The impact of the ZnS SEI layer on the EDL capacitance evolution was also studied by CV tests. Based on the equation C = εA/d, where ɛ is the dielectric constant of the electrolyte, A is electrode surface area and d is the EDL thickness, higher capacitance can be obtained with a higher specific surface area and smaller EDL thickness. As shown in Figs. S10 and 2f, the symmetric cell with Zn@ZnS electrodes shows much higher capacitance (156.68 μF cm⁻²) than that using bare Zn electrodes (60.30 μF cm⁻²), which is attributed to the abundant deposition sites on Zn@ZnS surface and decreased electric double layer. The higher capacitance is conducive to homogeneous zinc diffusion and deposition [1]. The enhanced electrokinetic and rich deposition sites can increase the reaction rate and decrease the reaction barrier, thus decreasing the potential drop over the compact layer and suppressing the diffuse layer.

It is well known that the solvation structure dominated by water is significant in determining the reaction kinetics since the hydrated Zn²⁺ needs to be de-solvated before gaining electrons [44]. The activation energy (E_a) is calculated based on the electrochemical impedance spectroscopy (EIS) tests. Figure 2g, h shows that with temperatures ranging from 50 to 80 ºC, the resistance of both electrodes decreases, and at a given temperature, the R_ct of the Zn@ZnS electrode (Fig. 2g) is much lower than that of the pure Zn anode (Fig. 2h), demonstrating an enhanced charge transfer capability. A semi-circle in a high-frequency area is also observed for the Zn@ZnS electrode, corresponding to the resistance associated with Zn²⁺ crossing the SEI layer. The calculated activation energy (E_a) based on the Arrhenius equation and the fitting results (Table S2) for the Zn electrode is 39.20 kJ mol⁻¹, in comparison to 30.89 kJ mol⁻¹ for the ZnS SEI layer (Fig. 2i). The decreased activation energy for Zn@ZnS cell can be attributed to the strong interaction between Zn²⁺ and electronegative sulfur elements weakens the repulsive force. What’s more, the water molecules in the solvation sheath are being attracted by the ZnS layer when diffused toward the zinc surface, advancing the desolvation process and the charge transfer rate. Additionally, the Zn²⁺ transference number tzn²⁺ was studied to illustrate the Zn²⁺ conductivity in the ZnS SEI layer (Fig. S11). The symmetric cell with bare Zn electrodes shows a low tzn²⁺ of only 0.15, while the cell with Zn@ZnS electrodes displays a high Zn²⁺ conductivity of 0.50. That’s because the hydrated Zn²⁺ moves slower than pure Zn²⁺ in an aqueous solution, and the lower desolvation rate causes the high concentration polarization of Zn (H₂O)₆²⁺ near the electrode vicinity. Meanwhile, SO₄²⁻ anions migrate toward the opposite electrode and result in charge separation, which limits the overall operating voltage because of the building of a junction potential. With the ZnS SEI layer, the decreased diffusion barrier and the fast de-solvation rate lowered the concentration gradient, and a fast Zn²⁺ conductance can be obtained.

The density functional theory (DFT) simulations were further performed to study the interaction between the absorbed Zn atom (adatom) and the Zn or Zn@ZnS electrode. As shown in Fig. 2j, the absorption energy of Zn- Zn@ZnS is − 2.21 eV (ZnS-Zn1 and ZnS-Zn2 represent different adsorption sites, see Fig. S12), much lower than the Zn-Zn interaction (− 0.26 eV), indicating the strong interaction between ZnS and adatom which enables more active sites for Zn²⁺ adsorption. The Bader charge analysis was then calculated to quantitively evaluate the electron transfer upon the adatom adsorption (Figs. 2k and S13). The results indicate with Zn-Zn adsorption, 0.045e was transferred from adatom to Zn in the substrate, while 0.180e was attracted by S, specifying the preferable and stronger adsorption of the adatom on the ZnS surface. The charge density difference in Figs. 2l and S14 intuitively shows the unbalanced charge distribution across the surface caused by the bonding interaction between the S and Zn adatom. The unbalanced charge distribution accelerates the Zn²⁺ diffusion across the ZnS layer and the deposition process because of the stronger electrostatic attraction toward Zn²⁺ other than the hydrated one in the electrical double layer, advancing the desolvation and reaction rate [31]. For zinc deposition at the Zn@ZnS electrode, there are energy barriers for the Zn²⁺ diffusing through the ZnS SEI layer to the zinc surface, which is 0.035 eV as calculated by DFT (Figs. S15 and S16). Therefore, shortening the diffusion path is required to achieve a fast reaction rate. In this regard, the ultrathin ZnS SEI layer with a thickness of 200 nm can minimize the energy barrier while ensuring a protective effect. As a result, the intrinsic affinity of ZnS toward Zn adatom enables more Zn²⁺ to diffuse through the ZnS film due to the decreased energy barrier and existing abundant nucleation sites, contributing to homogeneous Zn deposition and a dendrite-free morphology. Besides, to study the electrolyte surface in realistic environments, VASPsol was used to study the solvation energy of ZnS and Zn slab in water solvent conditions [45]. The results in Table S1 show that the Zn slab shows negligible solvation energy difference between vacuum and water environments; By contrast, the ZnS slab shows a more negative solvation energy of − 3.694 eV over a supercell of 7.62 × 9.90 Ų, implying a higher stabilization of the surface due to solvation [45].

To investigate the impact of regulated EDL on the battery’s performance, electrochemical characterizations were conducted to examine the suppression of side reactions and inhibition of zinc dendrite. The HER suppression was verified by the linear sweep voltammetry (LSV) test (Fig. 3a). The Zn@ZnS electrode requires higher HER overpotential (− 1.14 V vs. SCE) than the pure Zn electrode (− 1.0 V vs. SCE) for achieving the same reduction current density, indicating the HER is alleviated with the ZnS SEI layer [46]. The relieved zinc corrosion is also evidenced by the Tafel plot (Fig. 3b). With the SEI layer, the corrosion current decreases (1.19 vs. 1.33 mA cm⁻²), and corrosion potential increases from − 1.04 to − 1.02 V (vs. Ag/AgCl), suggesting the ZnS SEI layer can reduce the corrosion rate. The inhibited side reactions were further confirmed by XPS, X-ray powder diffraction (XRD), and SEM images. The XPS results of Zn@ZnS after cycling illustrate lower S content at 168.5 eV, corresponding to the S in Zn₄SO₄(OH)₆·H₂O (Fig. 3c). This implies the formation of the by-product is diminished with the ZnS protection layer, which is consistent with the XRD result and SEM images. Besides, the XPS results of Zn@ZnS after cycling also show the S 2p_1/2 (162.4 eV) and S 2p_3/2 (161.3 eV) peaks (Fig. S17), which corresponds to the S in ZnS state, indicating the stable existence of ZnS in the cycling process. The XRD results in Fig. S18 display that compared with the bare Zn electrode (0.91), the peak intensity ratio of the by-products and (002) plane (I_by-products/I₍₀₀₂₎) is lower for the Zn@ZnS electrode (0.5), proving limited by-products formation on Zn@ZnS anode after 100 h’s cycling. Figure S19 suggests that after soaking in 2 M ZnSO₄ (ZSO) for 5 days, a large flake-like by-product appears on the pure zinc surface, which was reported to be the Zn₄SO₄(OH)₆·H₂O. By contrast, the Zn@ZnS electrode surface shows a dense structure with only a few small flakes observed. These results indicate that the side reactions of HER and Zn corrosion are greatly inhibited with the protection of the ZnS SEI layer.

Coulombic efficiency (CE) was further examined in Zn//Cu asymmetric cells as a key parameter to reflect the reversibility of the battery running. As displayed in Fig. S20 at typical conditions of 1 mA cm⁻² and 1 mAh cm⁻², the CE of the cell using pure Zn anode fluctuates and drops in less than 90 cycles; by contrast, the Zn@ZnS||Cu cell cycles more than 220 cycles with an increasing CE, indicating the suppressed side reactions. The reduced voltage polarization (27.3 vs. 43.8 mV) and nucleation overpotential (44 vs. 56 mV) also confirm the decreased nucleation barrier enhanced reaction kinetics induced by ZnS. Increasing the current density to 5 mA cm⁻², the cell with Zn@ZnS anode manifests a high CE of 98.9% on average with good stability for more than 2,500 cycles while the CE of Zn||Cu cell fluctuated and failed at ~ 600 cycles (Fig. 3d). Besides, the charge/discharge curves of Zn@ZnS||Cu cell remain stable with a low voltage polarization of 56.7 mV (Fig. 3e), much lower than that using the Zn anode (85.97 mV) (Fig. 3f). This demonstrates that Zn@ZnS could constrain side reactions like the HER and notorious ‘‘dead Zn’’ formation, sustaining a high CE.

The impact of the ZnS SEI layer on the zinc dendrite suppression is theoretically studied by band structure calculation and experimentally verified by SEM images. The band structure of pure Zn, pure ZnS, and Zn@ZnS electrodes is calculated, as shown in Figs. S21 and 3g. The pure Zn metal shows no band gap, while the pure ZnS shows a band gap of 2.07 eV, consistent with the results reported before [47,48]. Compared with pure ZnS, the Zn@ZnS electrode interface still shows a semiconductor property but with a decreased band gap of 1.81 eV, that is because when the metal and the semiconductor are contacted, electrons from the conduction band of the semiconductor will flow into the metal until both Fermi levels equilibrate. With a semiconductor SEI layer, the zinc ions prefer to deposit under the ZnS film since the interface of Zn and ZnS has a lower potential. This is also proved by our EDX mapping results (Fig. S22). Figure S22 indicates that after 50 cycles, the Zn content in the surface area is lower than that in the bulk area because the zinc ions tend to deposit under the ZnS layer, which is conducive to the uniform deposition of zinc. The regulated zinc deposition with the ZnS SEI layer is further observed via the SEM. As depicted in Fig. 3h, i, the pure zinc surface after 50 cycles shows large flakes that are vertically oriented, potentially leading to the short-circuit of the battery. By contrast, the zinc surface with ZnS thin film presents a flatter surface and horizontally deposited zinc morphology, suggesting a dendrite-free zinc deposition. The uniform deposition of zinc is also confirmed by AFM images depicted in Fig. S23. The pure Zn anode shows a large average roughness (R_a) of 159 nm after 25 cycles, whereas a much flatter surface with Ra of 31.1 nm is obtained in the presence of the ZnS SEI layer. The different Zn deposition behavior is further studied by the chronoamperometry (CA) test (Fig. 3j). The classical nucleation theory (CNT) [49] suggests that the zinc nucleus is prone to aggregate during the deposition process to minimize the surface energy through the 2-dimension (2D) diffusion behavior. Then the following Zn²⁺ has a strong propensity to absorb on the preformed tips under the force of a higher electric field, leading to the zinc dendrite growth. Figure 3j shows that with an applied overpotential of − 150 mV, the Zn anode presents a continuous current density increase within 200 s, specifying a rampant 2D diffusion and vertical zinc deposition on the protuberances [50]. By contrast, the response current of the Zn@ZnS electrode stabilized in less than 20 s due to the strong interaction between S²⁻ and Zn²⁺, demonstrating a fast nucleation rate followed by a suppressed diffusion behavior.

To further understand the impact of ZnS-regulated EDL on the behavior of zinc plating, the finite element analysis performed by COMSOL Multiphysics was used to illustrate the zinc deposition behavior [51]. The zinc deposition and growth behavior were simulated at 5 mA cm^-2. Figure 3k illustrates that the electrolyte current density distribution is highly uneven across the Zn surface, with a significantly higher current near the protuberances than the flat surface area. Additionally, an intense distribution of the electric field can be observed near the protuberances (Fig. S22b). Consequently, the subsequent Zn²⁺ heavily diffuses and deposits around the nucleus pre-formed in the initial stage, leading to the formation of the zinc dendrite. On the contrary, with the SEI layer, the perturbance of the zinc surface was covered by the nanoparticle-like surface and a large surface area can be obtained. As shown in Figs. 3l and S24a, the Zn@ZnS electrode presents a uniform current density and electric field distribution, with a slightly higher current on the boundary of particles, facilitating the uniform Zn deposition and the formation of a flat surface. This is also confirmed by the electrodeposition thickness (Fig. S25). For the Zn electrode, most Zn deposits at the tips and little at the bottom flat area, while the Zn@ZnS electrode shows evenly distributed thickness, demonstrating the ZnS layer can alleviate the dendrite growth and promote uniform zinc deposition.

To further demonstrate the enhanced electrochemical performance of the Zn@ZnS anode, the galvanostatic charging/discharging process was studied in symmetric cells. Figure 4a exhibits the rate performance of the cells with a capacity of 1.0 mAh cm⁻². The cell with Zn@ZnS anode displays excellent cycling stability over various current densities, while the cell using pure Zn anode suffers a short circuit at 6.0 mA cm⁻². Besides, Fig. 4b indicates the cell with the ZnS protection layer shows much lower voltage hysteresis, especially at higher current density. That is because the good wettability of ZnS film decreases the surface energy and advances the Zn²⁺ diffusion toward the electrode. Besides, the zincophilicity ZnS layer can greatly reduce the diffusion barrier of hydrated zinc ions and promote the desolvation process. Such low voltage hysteresis induced by the ultrathin layer outperforms most previously reported coatings or SEI layers (Fig. 4c), revealing the great potential of the Zn@ZnS anode to achieve fast and deep energy storage.

The long cycling stability of the symmetric cells at various current densities and areal capacities was also carried out to investigate the suppressed zinc dendrite and side reactions induced by the optimized EDL. At typical conditions of 1 mA cm⁻² and 1 mAh cm⁻², the Zn@ZnS cell can stably cycle for almost 3,000 h, nearly 15 times longer than that of pure zinc anode (Fig. 4d). Besides, the voltage hysteresis of Zn@ZnS is more stable and much lower than the Zn anode (43.6 vs 81.1 mV) (Fig. S26). With an elevated current density of 3 mA cm⁻², the battery with Zn@ZnS anode exhibited long cycling for more than 1,500 h, while the battery using pure Zn anode failed after 250 cycles resulting from the internal short circuit caused by zinc dendrite growth (Fig. 4f). At a high current density of 5 mA cm⁻² and a capacity of 2 mAh cm⁻² (Fig. S27), the pure Zn cell only cycles for 70 h before the short circuit; By contrast, the battery with Zn@ZnS anode presents a stable running state for more than 350 h along with a lower voltage hysteresis, indicating the faster reaction kinetics and uniform plating of zinc deposits. To evaluate the performance under harsher practical conditions, the cycling performance was studied with a thin zinc foil of 20 μm. At 2 mA cm⁻² and 0.5 mAh cm⁻², the battery with ZnS SEI layer can stably cycle for around 600 cycles, in contrast to 150 cycles with pure Zn anode (Fig. S28). At a high depth of discharge (DOD) of 34.3%, the cell with Zn@ZnS anode still sustains a stable performance over 200 h, whereas the cell using pure Zn anode displays a fluctuating voltage and a short working time of 25 cycles (Fig. 4g). The electrochemical performance of the Zn@ZnS electrodes prepared at different reaction times and current densities were also tested at 2 mA cm⁻² and 2 mAh cm⁻² (Fig. S29). The results indicate the quality of the ZnS SEI layer largely depends on the charging capacity during the electrodeposition process, and a moderate charging capacity is decisive to getting a ZnS film with good protective effects. For instance, at a low current density of 1.25 mA cm⁻², the ZnS prepared under a longer deposition time of 8 min shows the best performance, while for a high current density of 5 mA cm⁻², a shorter deposition time of 1 min achieves the best performance. Overall, the longest lifetime is obtained at a combination of 2.5 mA cm⁻² and 2 min due to the uniform distribution of nanoparticle ZnS. Figure 4e compares the relationship between the lifetime and coating thickness. Ideally, a long battery life with a thinner coating layer is always desired since the inactive coating could reduce the energy density of the battery. Compared with the previously reported materials, the ultrathin ZnS SEI layer exhibits a competitive advantage in terms of long cycling performance, revealing the great potential of the Zn@ZnS anode to achieve an ultralong dendrite-free deposition behavior [19,28,30,36,37,40,52,53].

To test the practical application of the Zn@ZnS anode in ZIBs, the full cells with I₂/AC and MnO₂ as cathode materials were assembled. Figure S30a shows the XRD patterns of I₂ crystal (JCPDS#72-0012), active carbon (AC), and a composite of I₂/AC. The XRD peaks of the I₂/AC composite generally match with the XRD peaks of active carbon (AC), both exhibit two broad peaks at around 22° and 43°, corresponding to the (002) and (100) diffraction patterns of amorphous carbon materials. However, no evident I₂ peaks can be detected in the I₂/AC, indicating the formation of non-crystalline iodine [54]. Figure S30b indicates the α-MnO₂ matches with the PDF card of JCPDS#72-1982. Figure 5 shows the electrochemical performance of Zn||I₂@AC supercapattery and Zn||MnO₂ battery. The CV curves in Fig. 5a demonstrate that the zinc-iodine supercapattery with Zn@ZnS anode reached a higher current density and smaller redox overpotential due to the faster desolvation process and Zn²⁺ migration rate. This is consistent with the Nyquist plots of the cell at a different resting time shown in Fig. 5b. With the ZnS protection layer, the cells all displayed lower charge transfer resistance at the beginning and after 5 h’ resting, further proving the suppressed side reactions and enhanced interaction of Zn@ZnS anode with Zn²⁺ ions. The galvanostatic charging/discharging process was further conducted to elucidate the impact of the ZnS SEI layer in a full cell. The rate performance in Fig. 5c shows that the full cell with Zn@ZnS anode presents better rate performance and higher specific capacity, and a higher capacity of 160 mAh g⁻¹ is reached at 0.1 A g⁻¹. The charge and discharge curves of Zn@ZnS||I₂@AC full cell in Fig. 5d display a wide working window of 0.2-1.8 V and evident working plateaus at ~ 1.2 V, corresponding to the redox potential of I₂/I⁻ [55]. Figure S31 shows the long-term cycling stability at 10 A g⁻¹ of the cells with and without the ZnS layer. As reported, the generation of soluble polyiodide intermediates during the cycling process tends to corrode the zinc anode due to the “shuttle effect” [56]. Herein, to study the influence of the ZnS SEI layer on the inhibition of “shuttle effects”, the cell was subjected to 2,000 cycles followed by a day rest before restarting. Figure S31 shows that during the whole cycling process, the cell with Zn@ZnS anode exhibits higher capacity than the cell using a pure zinc anode. And a high capacity of 115 mAh g⁻¹ is obtained in the first 2,000 cycles. After 1 day of rest, the cell can still stably cycle for more than 8,000 cycles, with a high capacity retention of 82%. By contrast, the cell with pure zinc anode exhibits a faster capacity decay after one day’s resting, and the capacity drops to 60 mAh⁻¹ after 10,000 cycles. To meet harsh operating conditions, the long cycling performance of full cells with a low N/P ratio of 3 (ultrathin zinc foil of 20 μm) was further studied at a current density of 1A g⁻¹ (Fig. 5e). With the ZnS protection layer, the full cell sustains stable cycling for more than 900 cycles, with a capacity retention of 63%. In comparison, the cell with pure zinc anode manifests lower specific capacity than the Zn@ZnS anode in the first 400 cycles and then suffers drastic capacity decay to zero. These results suggest that the side reactions and zinc dendrite are greatly suppressed with ZnS coating.

The electrochemical performances of Zn/MnO₂ full cells were further studied. The cell with Zn@ZnS anode shows a better rate performance with a higher specific capacity over various current densities (Fig. 5f), and a higher capacity of 250 mAh g⁻¹ is obtained at 0.1 A g⁻¹, compared to 220 mAh g⁻¹ for the bare Zn. This is consistent with the results obtained with I₂/AC cathode. The higher capacity of full cells in the first few cycles with Zn@ZnS anode can be attributed to the fast reaction kinetics and suppressed side reactions. The charge and discharge curves in Fig. S32 display two typical discharge plateaus caused by consequent H⁺/Zn²⁺ intercalation [57]. Importantly, with Zn@ZnS anode, the cell sustains both higher capacity and long cycling stability of 130 mAh g⁻¹ after 1,200 cycles at 0.5 A g⁻¹, while the cell with bare Zn under a fast capacity decay within 600 cycles (Fig. 5g). The enhanced performance can be attributed to the following aspects. Firstly, the suppressed side reactions caused by the ZnS SEI layer protect the anode from corrosion and alleviate the cathode dissolution due to the inhibition of water decomposition. Secondly, the good wettability and the enhanced zinc ions migration contribute to the low-concentration polarization and high reactivity in the ZnS-modified battery.