Electrolyte additive screening for co-regulation of solvation and solid electrolyte interphase in aqueous zinc batteries Authors: Shaocong Tang, Jiabao Li, Tian Wang, Tianyi Wang, Weiwei Xiang, Ya Xiao, Hong Mu Park, Shenqiu Xu, Jae Su Yu, Huan Pang Published: 2026 ABSTRACT AbstractThe development of aqueous zinc metal batteries is hindered by dendritic growth, parasitic side reactions, and unstable electrode-electrolyte interface, leading to rapid capacity fading. To address these issues, we rationally screen and identify a pair of complementary additives, methyl propionate and L-cysteine, to synergistically regulate zinc ion solvation structure and interfacial chemistry. Methyl propionate primarily lowers the energy barrier for zinc ion desolvation by weakening the interaction between zinc ions and water molecules, while L-cysteine stabilizes the interface by forming a protective and inorganic-rich solid electrolyte interphase through strong adsorption and preferential reduction. Besides, the incorporation of methyl propionate and L-cysteine reorganize the hydrogen-bond network, lower water activity, and suppress parasitic reactions. Consequently, the optimized electrolyte enables Zn||Zn symmetric cells to deliver stable plating/stripping over 500 h at 10 mA cm−2 and 10 mAh cm−2. This work demonstrates a molecular-level design strategy that integrates solvation regulation with solid electrolyte interphase engineering, providing insights into the rational design of aqueous zinc metal batteries.Subject terms: Batteries, Batteries Introduction Aqueous zinc (Zn) metal batteries (AZMBs) have attracted considerable attention as a promising energy storage system due to their high theoretical capacity (820 mAh g−1), intrinsic safety, and cost-effectiveness. Unlike lithium-ion batteries that rely on flammable organic electrolytes, AZMBs utilize aqueous electrolytes, reducing safety risks and environmental concerns.1,2 Additionally, Zn metal is earth-abundant and exhibits a low redox potential (−0.76 V vs. standard hydrogen electrode), making it a highly competitive electrode material for large-scale applications such as grid energy storage and flexible electronics.3–5 However, despite these advantages, the practical deployment of AZMBs remains limited by several fundamental challenges, including dendrite growth, hydrogen evolution reaction (HER), and interfacial instability, which result in rapid capacity degradation, poor Coulombic efficiency (CE), and limited cycling reversibility.6–8 Specifically, these challenges primarily originate from the intrinsic interactions among Zn2+, water molecules, and the electrode interface. In conventional ZnSO4 (ZS) electrolytes, Zn2+ predominantly exists as the [Zn(H2O)6]2+ complex, where six water molecules coordinate in an octahedral geometry.9 While this solvation shell ensures Zn2+ stability in bulk electrolytes, it introduces several electrochemical limitations: 1) strong Zn2+-H2O interactions impose a high desolvation energy barrier, slowing Zn2+ reduction and leading to inhomogeneous ion distribution, which promotes dendritic deposition; 2) the accumulation of non-coordinated water molecules at the Zn interface facilitates HER and Zn corrosion, thereby compromising interfacial stability and lowering CE; 3) reductive decomposition of the electrolyte forms a structurally unstable solid electrolyte interphase (SEI) with limited Zn2+ transport capability, further aggravating interfacial side reaction and non-uniform Zn deposition.10,11 In general, the Zn2+ solvation structure, HER activity, and SEI functionality are inherently governed by the electrolyte environment. To address these issues, a rationally designed electrolyte that simultaneously optimizes Zn2+ coordination, suppresses HER, and stabilizes the Zn/electrolyte interface is essential for achieving a highly stable Zn metal electrode. Electrolyte engineering strategy typically includes solvent design, salt optimization, and additive incorporation. Among them, electrolyte additives have emerged as an effective strategy for regulating solvation structure, interfacial composition, and suppressing HER in AZMBs. According to recent relevant studies, electrolyte additives can be divided into two main types based on their functional mechanisms. Small polar organic molecules represent one class of functional additives that can partially replace H2O in the primary Zn2+ solvation sheath via moderate binding affinity toward Zn2+, which reduces the overall stability of the Zn2+ solvation shell, thereby weakening Zn2+-H2O interactions.12–14 Moreover, these molecules interact with surrounding water via hydrogen-bonding (H-bonding), disrupting the intrinsic H2O-H2O hydrogen-bond (H-bond) network. This combined effect lowers the desolvation energy barrier and facilitates uniform Zn deposition.15,16 Similar trends have been reported in systems employing N-Methylpyrrolidone,17 dimethyl sulfoxide,18 or ether-based additives,19 where the weaker Zn2+ coordination leads to reduced desolvation energies and improved Zn plating kinetics. However, despite the extensive progress of organic solvents in modifying the solvation sheath, they exhibit limited interfacial affinity toward the Zn surface, which restricts their ability to form stable and protective SEI during long-term cycling. In contrast, biomolecules such as amino-acid-based additives represent a class of interfacial regulators that interact strongly with the Zn surface through multiple functional groups (e.g., -NH2, -COOH, and -SH).20,21 These groups enable preferential adsorption onto the Zn surface, and their intrinsically low lowest unoccupied molecular orbital (LUMO) levels facilitate electrochemical reduction, leading to the formation of inorganic-rich SEI components that effectively suppress parasitic reactions and improve interfacial stability. However, a limitation is that amino acids contain multiple electron-donating groups and generally exhibit strong coordination ability toward Zn2+, which tends to further stabilize the solvation shell rather than facilitate desolvation. Notably, recent research focused on either solvation structure regulation or interfacial engineering, but such efforts were largely fragmented and lacked a unified design rationale. Moreover, these studies typically relied on empirical and trial-and-error selection of electrolyte additives, without robust theoretical guidance to support systematic optimization or deeper mechanistic understanding. In this work, unlike the majority of the previous studies that focused solely on either solvation structure modulation or interfacial regulation, we propose a unified framework that integrates both aspects to systematically guide the design of electrolyte additives. Importantly, instead of relying on empirical or trial-and-error methods, an additive screening framework guided by the complementary mechanisms of solvation modulation and interfacial stabilization was developed and applied to a series of candidate molecules, including organic and amino-acid species. Their Zn2+ coordination strength, H-bonding characteristics, and interfacial reactivity were systematically investigated using density functional theory (DFT) calculations. As a result, methyl propionate (MP) and L-cysteine (L-Cys) were identified as optimal additives for the regulation of Zn2+ solvation structure and interfacial chemistry. Their co-introduction induces a transformation of the Zn2+ solvation structure from [Zn(H2O)6]2+ to [Zn(H2O)4-MP-L-Cys]2+, which optimizes the Zn2+ desolvation process and significantly reduces the desolvation energy barrier. Molecular dynamics (MD) simulations confirmed that MP disrupts bulk water structure and facilitates the release of H2O ligands from the Zn2+ solvation sheath. Meanwhile, L-Cys preferentially adsorbs onto the Zn surface and undergoes reductive decomposition, contributing to the formation of a stable and inorganic-rich SEI composed of Zn3N2, ZnS, and ZnO. Besides, this interphase exhibits a more delocalized electronic structure and lower Zn2+ diffusion barriers compared to conventional Zn(OH)2 layers, supporting rapid interfacial ion transport and enhanced chemical stability. The synergistic interaction between MP and L-Cys enables concurrent optimization of bulk Zn2+ coordination and interfacial stability, resulting in highly reversible Zn plating/stripping with suppressed side reactions and extended cycling life in both Zn||Zn symmetric cells and Zn||I2 pouch cells. Results Electrolyte additive screening To validate the proposed dual-regulation screening framework targeting solvation structure and interfacial chemistry, distinct sets of molecular descriptors were employed for those two categories of additives, including 1,2-dimethoxyethane (DME), tetrahydrofuran (THF), ethylene carbonate (EC), propylene carbonate (PC), ethyl propanoate (EP), ethyl acetate (EA), dimethyl carbonate (DMC), methyl acetate (MA), MP, glycine (Gly), alanine (Ala), serine (Ser), aspartic acid (Asp), glutamic acid (Glu), histidine (His), methionine (Met), threonine (Thr), arginine (Arg), and L-Cys. For the solvation regulation, organic additives with moderate binding affinities to Zn2+ and H2O generally enable reconfiguration of the Zn2+ solvation shell and disruption of the H2O-H2O H-bond network, thereby facilitating Zn²⁺ desolvation.16 Meanwhile, a low electrostatic potential difference (ΔESP) helps suppress proton-induced side reactions (Fig. 1a–c, Supplementary Figs. 1 and 2, and Supplementary Data 1–29). However, the organic solvents exhibit weak interaction with the Zn (002) surface, indicating that they alone are insufficient to induce the formation of a stable SEI layer. In contrast, amino acid molecules generally exhibit strong surface affinity, indicating their potential for interfacial modulation (Fig. 1d, Supplementary Fig. 3 and 4 and Supplementary Data 30–49). Accordingly, their ability to promote interfacial reduction and facilitate robust SEI formation was further evaluated based on their adsorption energy and LUMO levels on the Zn(002) surface (Fig. 1d–f and Supplementary Data 50–59). Electrolyte screening based on DFT and MD. a Electrostatic potential (ESP) distributions of various organic molecules. b Binding energies of various organic molecules with Zn2+ (blue) and H2O (pink). c Organic molecules screening based on binding energy with H2O, binding energy with Zn2+, and ΔESP. d Adsorption energy comparison of different molecules on the Zn (002) facet. e Highest occupied molecular orbital (HOMO) and LUMO levels of different molecules. f Amino acids screening based on adsorption energy, LUMO level, and Zn2+ binding energy. Model of the g ZS and h ZSLM electrolytes utilized in the MD simulations. RDF results from the MD simulations of the i ZS and j ZSLM electrolytes. Source data are provided as a Source Data file. Based on the above screening strategy, six representative molecules, including MP, L-Cys, DME, PC, Asp, and Ser were selected, covering a wide range of coordination strength, interfacial affinity, and electronic structure. Among these molecules, MP, DME, and PC represent organic solvent-type additives with varying Zn2+ and H2O binding affinities. Among them, MP was chosen for its moderate Zn2+ and H2O coordination and relatively low ΔESP, which collectively enable moderate solvation shell reconfiguration and disruption of the H-bond network, thereby facilitating Zn2+ desolvation and suppressing proton-induced side reactions. For DME, it exhibits the lowest ΔESP, which can also mitigate proton-induced side reactions; however, its weak interactions with Zn2+ and H2O limit its ability to modulate the solvation structure. In contrast, PC exhibits strong coordination with both Zn2+ and H2O, along with the highest ΔESP, which may lead to over-stabilization of the solvation shell and increased proton affinity, thus hindering desolvation and increasing the risk of parasitic side reactions. This selection enables systematic assessment of how additives interact with Zn2+ and H2O influence solvation shell restructuring, Zn2+ desolvation kinetics, and the suppression of side reactions. Additionally, given the limited interfacial activity of organic solvents, L-Cys, Asp, and Ser were selected as representative amino acid additives to systematically investigate this interfacial behavior. L-Cys was selected for its strong adsorption energy, low LUMO level, and relatively weak Zn2+ binding energy among these amino acids, which facilitate interfacial reduction and robust SEI formation without over-stabilizing the solvation shell (Supplementary Fig. 5 and Supplementary Data 60–69). While Asp shows relatively weak surface adsorption and the highest LUMO level, suggesting a limited capacity to promote interfacial reconstruction. For Ser, it possesses the strongest adsorption energy and the lowest LUMO level, which may lead to high desolvation energy or high interfacial resistance. This selection enables systematic evaluation of how variations in interfacial affinity and electronic structure affect Zn surface chemistry and interphase stability. To experimentally verify the solvation and interfacial trends predicted by the theoretical screening, we conducted 1H nuclear magnetic resonance (NMR) spectroscopy, Fourier-transform infrared (FTIR), and Raman spectroscopy on the eight corresponding electrolyte systems.22–24 The 1H NMR spectra show that the ZS + PC (ZSP) electrolyte exhibits the most pronounced upfield shift, followed by ZS + MP (ZSM), while ZS + DME (ZSD) and amino-acid-containing systems show minimal shifts, reflecting differences in Zn2+-H2O interaction and the ability of additives to reconstruct the solvation shell and H-bond network (Supplementary Fig. 6a and Supplementary Fig. 7a). Unlike MP that weakens Zn2+-H2O interactions by disrupting the H-bond network and modulating the solvation shell to facilitate desolvation, PC strongly coordinates with Zn2+, which over-stabilizes the solvation structure and hinders Zn2+ desolvation, especially when combined with amino acid additives. In contrast, the small chemical shifts in amino-acid systems are attributed to interfacial dominant effects with limited impact on bulk solvation. FTIR spectra further reveal that ZSP exhibits a more pronounced O-H blue-shift than ZSM (Supplementary Fig. 6b and Supplementary Fig. 7b), consistent with the trend observed in 1H NMR. In Raman spectra (Supplementary Fig. 8a), the SO42− symmetric stretching band near 980 cm−1 blue-shifts in ZSP and ZSM, suggesting tighter ion pairing due to solvation reorganization. Furthermore, the deconvolution of the O-H stretching region (2800–4000 cm−1) shows that in ZSP, ZSM, and ZS + L-Cys + MP (ZSLM), the fraction of strong H-bonds decreases significantly, while the content of weak H-bonds increases, indicating that the H-bond network was disrupted in these electrolytes. In contrast, ZS + L-Cys (ZSL), ZS + Asp (ZSA), and ZS + Ser (ZSS) maintain relatively strong H-bond content, consistent with their interfacial dominant behavior (Supplementary Fig. 8b). In particular, the favorable chemical shift in ZSLM reflects the complementary roles of MP and L-Cys in modulating solvation and interfacial chemistry, thereby validating the reliability of the screening framework. Electrochemical performance in Zn||Zn symmetric cells further validates the predicted solvation and interfacial behaviors, with ZSLM exhibiting the lowest nucleation overpotential and most stable Zn plating/stripping, confirming MP and L-Cys as ideal complementary additives for synergistic regulation of solvation structure and interfacial chemistry (Supplementary Figs. 9 and 10 and Supplementary Notes 1 and 2). While thermodynamic and electronic descriptors confirmed the individual potential of MP and L-Cys in tuning both solvation and interfacial stability, a detailed understanding of their synergistic effects necessitated further structural investigation. MD simulations were performed to analyze how the dual-additive modulates the Zn2+ solvation structure. In the ZS, Zn2+ is primarily coordinated with six H2O to form a highly hydrated [Zn(H2O)6]2+ complex (Fig. 1g and Supplementary Data 70 and 71). The introduction of MP and L-Cys fundamentally reconstructs the solvation structure, yielding a [Zn(H2O)4-MP-L-Cys]2+ configuration where both additives partially replace coordinated H2O (Fig. 1h and Supplementary Data 72 and 73). Radial distribution functions (RDF) analysis confirms this shift, with the Zn2+-H2O coordination number (CN) decreasing from 6.17 in ZS to 3.83 in ZSLM, while L-Cys and MP contribute CN of 1.08 and 0.87, respectively (Fig. 1i, j). Given the co-coordination of MP and L-Cys in the ZSLM, their individual coordination behaviors were further examined in the ZSL and ZSM electrolytes. In ZSL, RDF analysis shows that L-Cys participates in Zn2+ coordination with a CN of 1.04, reducing H2O coordination to 4.78 (Supplementary Fig. 11 and Supplementary Data 74 and 75). This is attributed to the functional groups of L-Cys, which promote stable coordination with Zn2+. While MP in the ZSM system exhibits moderate coordination strength (CN = 0.93), with H2O coordination maintained at 5.07 (Supplementary Fig. 12 and Supplementary Data 76 and 77). Such moderate coordination indicates a weaker but sufficient interaction to perturb the hydration structure, lowering desolvation energy barrier. Notably, the solvation structure affects the desolvation energy barriers of Zn2+. Supplementary Figs. 13–16 reveal distinct step-by step desolvation energy for the different electrolytes. In ZSLM, the energy barriers for step-by-step H2O dissociation are significantly lower than those observed in the ZS electrolyte. This behavior originates from the weak coordination between MP and Zn2+ in ZSLM, promoting the initial desolvation process and reducing the desolvation energy barrier. This trend is further supported by the lowest energy barriers exhibited in the ZSM electrolyte. In contrast, the ZSL electrolyte exhibits the highest desolvation energy barrier, which can be attributed to the multidentate coordination of L-Cys via -NH2, -COOH, and -SH.21 This strong coordination enables L-Cys to maintain stable binding with Zn2+ during the later stages of desolvation, acting as a chemical precursor for interfacial reactions and thereby promoting the formation of an inorganic-rich SEI. To gain further insight into the distinct roles of MP and L-Cys, we evaluated the wettability of ZS, ZSM, ZSL, and ZSLM electrolytes by measuring the contact angle on Zn foils. As shown in Supplementary Fig. 17, the contact angle progressively decreases from ZS to ZSM to ZSL, reaching the lowest value in ZSLM, indicating that the dual-additive system significantly enhances Zn-electrolyte interfacial wettability, which contributes to the reduced nucleation overpotential and more uniform Zn deposition.25 As shown in Supplementary Fig. 18, after 100 cycles, the Zn negative electrode still achieves smooth and compact deposition without dendritic growth or agglomeration in ZSLM electrolyte, in contrast to severe structural degradation in ZS, flake-like agglomerates in ZSM, and fine dendrites in ZSL.26 In addition, ionic conductivity and viscosity measurements demonstrate that MP markedly improves ionic conductivity and lowers viscosity by disrupting the H-bond network and weakening water-water interactions, thereby facilitating Zn2+ transport in the bulk electrolyte, while L-Cys has a slight influence on these bulk properties (Supplementary Fig. 19). These findings are supported by diffusion coefficients and transference numbers of Zn2+. The mean square displacement of Zn2+ increases in all the additive-containing systems compared to the ZS electrolyte, with the ZSLM electrolyte exhibiting the highest diffusion coefficient of 9.33 × 10−6 cm2 s−1 (Supplementary Fig. 20). Consistently, the Zn2+ transference number increases from 0.39 in ZS to 0.60 in ZSL and 0.68 in ZSM, reaching 0.79 in ZSLM (Supplementary Fig. 21). These results highlight a synergistic mechanism of MP and L-Cys, jointly promoting homogeneous ion flux and dendrite-free Zn deposition. HER suppression In addition to optimizing Zn2+ solvation and interfacial chemistry, suppressing the HER is equally critical for improving the electrochemical reversibility. To systematically elucidate the impact of the dual-additive on the H-bond network and its correlation with HER suppression, MD simulations were performed to visualize and quantify H-bonding interactions across different electrolyte environments (Fig. 2a, b, and Supplementary Figs. 22–26). In the ZS electrolyte, the number of H2O-H2O H-bond reaches 13.6 × 103, indicating a highly interconnected water network (Supplementary Fig. 23a). In the ZSL system, 1.4 × 103 H2O-L-Cys H-bonds are formed, while the number of intrinsic H2O-H2O H-bonds slightly decreases to 11.7 × 103 (Supplementary Fig. 23b), which results from the stronger H2O-L-Cys binding energy (−8.78 kcal mol−1) compared with that between two H2O molecules (−5.48 kcal mol−1), thereby leading to the participation of L-Cys in H-bonding with water (Supplementary Fig. 24). In contrast, MP significantly alters the H-bonding environment in the ZSM electrolyte, forming 2.7 × 103 H2O-MP H-bonds while reducing H2O–H2O H-bonds to 9.9×103 (Supplementary Fig. 23c), which reflects a much stronger capacity to reorganize the H-bond network. It is worth noting that this difference arises from their distinct functional roles in the electrolyte (Supplementary Fig. 27). L-Cys, owing to its strong adsorption on the Zn (002) surface, tends to anchor at the Zn surface, where it thereby integrates into the H-bond network near the Zn negative electrode through its multiple H-bond donor and acceptor groups (-COOH, -NH2).27 Conversely, MP exhibits weak adsorption on the Zn surface, suggesting its tendency to remain in the bulk electrolyte, where its polar C = O group serves as a directional H-bond acceptor that disrupts the H2O-H2O network and anchors water molecules through specific interactions, thereby reducing its activity. Thus, the co-introduction of MP and L-Cys further reduces the number of H2O-H2O H-bonds to 8.5 × 103, accompanied by 2.4 × 103 H2O-MP and 1.0 × 103 H2O-L-Cys H-bonds in the electrolyte, respectively (Fig. 2b). This cooperative reconstruction of the H-bond network highlights the synergistic modulation of bulk solvation and interfacial environments achieved by MP and L-Cys. H-bond network regulation suppresses hydrogen evolution. a Snapshots of the H-bond network in ZSLM. b Number of different H-bonds in ZSLM electrolytes. c ΔGH* for the different electrolytes on Zn surface. d Tafel plots of the Zn negative electrode tested with a three-electrode system in different electrolytes. e Linear sweep voltammetry (LSV) curves of Zn plate tested with a three-electrode system in different electrolytes. Hydrogen diffusion flux (f) in ZS electrolyte and g in ZSLM electrolyte. Source data are provided as a Source Data file. In addition, the effects of MP and L-Cys on HER were evaluated by calculating the Gibbs free energies of H+ adsorption (ΔGH*) at the Zn interface (Fig. 2c and Supplementary Fig. 28). In the ZS electrolyte, ΔGH* is calculated to be 0.66 eV, indicating that hydrogen adsorption is thermodynamically favorable, which facilitates HER. This can be attributed to the high concentration of free water, which provides abundant protons for HER, exacerbating Zn corrosion and interfacial instability.28 In contrast, in the ZSLM electrolyte, ΔGH* increases significantly to 1.67 eV, indicating a much higher energy barrier for hydrogen adsorption and a substantial reduction in HER activity, which can be attributed to the presence of L-Cys and MP, which modulates the H-bond environment, reducing the active hydrogen adsorption sites, thereby suppressing hydrogen evolution. The interfacial corrosion behavior of Zn negative electrodes in ZS, ZSM, ZSL, and ZSLM electrolytes was further examined by immersing Zn foils for various durations. As shown in Supplementary Fig. 29, scanning electron microscope (SEM) images of Zn foils soaked in the ZS electrolyte for 7 and 15 days show severe surface degradation with the accumulation of a large amount of by-products. Although the ZSM and ZSL electrolytes partially alleviate corrosion, with increasing soaking time, the Zn surface in ZSM becomes gradually covered with flake-like corrosion products, while the Zn surface in ZSL develops unevenly distributed corrosion spots or aggregated regions. In contrast, even after 15 days of soaking, the Zn surface in the ZSLM electrolyte remains clean and smooth, without observable by-products. In addition, the X-ray diffraction (XRD) patterns reveal that Zn foils soaked in ZS, ZSM, and ZSL electrolytes exhibit diffraction peaks of Zn4SO4(OH)6·5H2O, ZnSO3·2H2O, and Zn3O(SO4)2 after both 7 and 15 days of immersion, indicating the formation of by-products on the Zn surface (Supplementary Fig. 30). By comparison, these by-product peaks are almost absent on Zn soaked in the ZSLM electrolyte, indicating that ZSLM effectively inhibits corrosion-induced interfacial degradation and enhances the long-term stability of Zn negative electrode.29 To further validate the suppression of water-induced side reactions, the corrosion resistance of Zn negative electrode in different electrolytes was evaluated using linear polarization measurements (Fig. 2d). The corrosion current density of the Zn negative electrode in the ZS electrolyte was measured to be 0.897 mA cm−2, indicating a high corrosion tendency. Upon the introduction of MP and L-Cys, the corrosion current density significantly decreased to 0.164 mA cm−2 in the ZSLM electrolyte, suggesting an improved resistance to Zn corrosion. Furthermore, HER activity was assessed through HER onset potential measurements (Fig. 2e). The Zn negative electrode in the ZSLM electrolyte showed the lowest HER onset potential compared with the other three electrolytes, further confirming that the side reactions caused by H2O were greatly suppressed. Finite element simulation (FES) further reveals the HER suppression mechanism. In the ZS electrolyte, H+ rapidly accumulates at surface protrusions due to localized electrostatic effects, leading to severe hydrogen evolution (Fig. 2f).30 In comparison, both ZSM and ZSL exhibit reduced H⁺ accumulation near the Zn surface, indicating alleviated proton enrichment and improved interfacial stability (Supplementary Figs. 31 and 32). Notably, the ZSLM electrolyte exhibits the lowest interfacial H+ concentration among all systems, reflecting a more uniform ionic flux and effectively suppressed HER (Fig. 2g). Zn deposition behavior Surface passivation caused by water-induced side reactions and uncontrolled dendrite growth on the Zn negative electrode severely compromises the reversibility of AZMBs. Given that the ZSLM electrolyte exhibits the most comprehensive improvements in ion transport kinetics, Zn deposition behavior, interfacial stability, and desolvation thermodynamics among the four electrolytes (Supplementary Fig. 33 and Supplementary Table 1), we further investigated the synergistic regulation of Zn deposition by the dual-additive, with particular attention to the evolution of deposition morphology and interfacial dynamics. XRD analysis (Fig. 3a) reveals that the electrode cycled in the ZSLM electrolyte exhibits a more pronounced Zn (002) peak compared to that in the ZS electrolyte. Although Zn (101) remains the dominant diffraction peak due to its intrinsic intensity in the hexagonal close-packed lattice, the increased intensity ratio of Zn (002) to Zn (101) (I (002):I (101) = 0.74) in the dual-additive system indicates a preferred orientation along the (002) plane, which implies that Zn preferentially deposits along the thermodynamically stable (002) facet, resulting in suppressed dendrite formation.30 Furthermore, the ZS electrolyte exhibits additional diffraction peaks corresponding to byproducts after cycling, whereas these signals are absent in the ZSLM electrolyte, indicating that the dual-additive electrolyte effectively suppresses parasitic reactions and inhibits the formation of insoluble hydroxide by-products, thereby maintaining interfacial stability during long-term cycling. To further probe the Zn2+ interfacial behavior associated with this structural evolution, chronoamperometry (CA) measurements were performed (Fig. 3b). The ZSLM electrolyte rapidly reaches a steady-state current, consistent with a three-dimensional (3D) diffusion-controlled nucleation process.31 In contrast, the ZS electrolyte displays a continuous current decay, indicative of unregulated 2D nucleation and disordered Zn2+ diffusion.32 This observation suggests that the dual-additive system facilitates a favorable interfacial environment that promotes uniform Zn nucleation and suppresses the formation of dendrites. The Zn2+ transport kinetics at the Zn/electrolyte interface were evaluated through electrochemical impedance spectroscopy (EIS). Based on the Arrhenius equation,33 the calculated activation energy (Ea) for the ZSLM electrolyte is significantly reduced to 24.61 kJ mol−1, compared to 36.42 kJ mol−1 for the ZS electrolyte (Fig. 3c and Supplementary Fig. 34), confirming that the dual-additive strategy lowers the desolvation energy and accelerates interfacial charge transfer, which is in agreement with the results of DFT calculation. Zn deposition behavior. a XRD patterns of Zn negative electrode after cycling in ZS and ZSLM electrolytes. b CA curves of Zn negative electrode using ZS and ZSLM electrolytes. c Arrhenius curves and comparison of activation energies. AFM images of Zn negative electrodes after 50 cycles in d ZS and e ZSLM electrolytes. In-situ OM images after Zn plating: f in ZS and g in ZSLM electrolytes. Operando Raman spectra in h ZS and i ZSLM electrolytes. j, k Dual-field simulations of the electric field and Zn2+ concentration distribution disparities in ZS and ZSLM electrolytes. Source data are provided as a Source Data file. To investigate how the enhanced interfacial Zn2+ transport affects Zn deposition morphology, the surface structure of the cycled Zn negative electrode was investigated by atomic force microscopy (AFM). As shown in Fig. 3d, e and Supplementary Fig. 35, the Zn negative electrode cycled in the ZS electrolyte exhibits a rough and heterogeneous surface, whereas the surface becomes notably more uniform and flatter upon the introduction of L-Cys and MP. Confocal laser scanning microscopy (CLSM) was additionally employed to provide a more comprehensive view of the Zn morphology after cycling.34 The Zn negative electrode cycled in ZSLM exhibits a smooth and uniform surface with an average roughness (Sa) of only 9.52 μm (Supplementary Fig. 36a, b), while that in ZS presents pronounced surface roughness (Sa = 46.30 μm) and evident dendritic structures (Supplementary Fig. 36c, d), indicating severe morphological degradation. This confirms that the co-regulation by MP and L-Cys enables uniform Zn nucleation and suppresses dendritic growth, thereby maintaining interfacial stability over extended cycling. In addition, in-situ optical microscopy (OM) was employed to directly visualize Zn deposition behavior. As shown in Fig. 3f, the ZS electrolyte exhibits pronounced bubbles and uncontrolled dendrite growth within 50 min, confirming the occurrence of HER and uneven Zn deposition. In contrast, the ZSLM electrolyte enables smooth and homogeneous Zn deposition throughout the whole period of 50 min electroplating process (Fig. 3g). In addition, as shown in Supplementary Fig. 37 and Fig. 3h, i, in-situ Raman spectroscopy was used to monitor Zn2+ transport at the electrolyte-electrode interface. The Raman signal at ~980 cm−1 corresponds to SO42−, which is applied to measure variations in Zn2+ concentration during the electroplating process. In the ZS electrolyte, the signal shows a significant SO42− depletion, indicating that Zn2+ deposition occurs rapidly and unevenly. On the contrary, in the ZSLM electrolyte, the peak exhibits a more stable and gradual change in intensity, suggesting that Zn2+ consumption is significantly alleviated during electrodeposition, thereby avoiding concentration-induced dendrite formation.35 To gain mechanistic insight into how interfacial regulation governs Zn deposition morphology, FES was employed to probe the evolution of surface electric field and Zn2+ concentration profiles during cycling parameters that directly influence nucleation uniformity and dendrite suppression.36 In addition to a more homogeneous surface charge distribution compared to the ZS electrolyte, the ZSLM electrolyte also exhibits a stable and uniform Zn2+ flux (Fig. 3j, k). These features collectively alleviate interfacial charge accumulation, suppress the notorious “tip effect” associated with dendrite formation, and effectively regulate uniform Zn deposition. Structure and composition of the SEI layer The interfacial modulation induced by L-Cys additive was examined by time-of-flight secondary ion mass spectrometry (TOF-SIMS). The 3D rendering of TOF-SIMS (Fig. 4a) reveals the formation of a layered SEI film in the ZSLM electrolyte. Organic fragments such as CH− and CN− are primarily distributed in the outermost layer, while Zn-containing inorganic species such as ZnO−, ZnS−, and ZnN− are enriched in the inner SEI layer, indicating the construction of a chemically stratified bilayered SEI. This bilayer structure is further supported by 2D TOF-SIMS images and sputtering depth profiles (Supplementary Figs. 38a–c and 39a–c). The CH− and CN− signals show strong intensity at the surface but diminish rapidly with increasing sputtering time (Supplementary Fig. 38d, e), confirming their localization in the outer SEI layer. The depth profile of ZnOH−, a representative by-product of parasitic side reactions, exhibits minimal intensity throughout the sputtering process in the ZSLM electrolyte (Supplementary Fig. 38f), suggesting that hydroxide formation is effectively suppressed during SEI formation. In contrast, ZnN− and ZnS− intensities gradually increase (Supplementary Fig. 39d, e), indicating their accumulation in the inner SEI region. In comparison, the SEI formed in ZS electrolyte displays a disordered and chemically inhomogeneous structure. As shown in Supplementary Figs. 40 and 41, ZnOH− and ZnO− signals are both abundant and broadly distributed across the surface and interior of the SEI film, reflecting uncontrolled interfacial reactions and the absence of a well-defined SEI framework. The lack of functional stratification, along with the excessive accumulation of ZnOH−, leads to interfacial instability and ineffective regulation of Zn2+ flux, thereby accelerating dendrite formation and degrading electrochemical performance.37 To corroborate the bilayer SEI structure observed via TOF-SIMS, Ar+ sputtering X-ray photoelectron spectroscopy (XPS) and high-resolution transmission electron microscopy (HRTEM) were employed to further resolve its chemical distribution and spatial configuration at the molecular and nanometer scales. As shown in Fig. 4b–d and Supplementary Fig. 42, the C 1 s and N 1 s spectra reveal the presence of C-C/C-H species, indicating the in-situ formation of an organic-rich SEI layer on the Zn negative electrode in ZSLM electrolyte.38 As the sputtering time increases, the intensities of these organic signals gradually decrease and even disappear, implying their predominant distribution in the outer SEI layer. In contrast, the N 1 s, S 2p, and O 1 s spectra exhibit progressively stronger signals associated with N-Zn, ZnS, and O-Zn, suggesting that the inorganic components are mainly concentrated in the inner region of the SEI.30,39,40 Meanwhile, the CO32− signal intensity also increases with sputtering depth, which is attributed to the reaction between early-formed alkaline species such as Zn(OH)2 and CO32−/HCO3− generated from dissolved CO2 and H2O in the electrolyte.37 This bilayer SEI layer is further confirmed by HRTEM (Fig. 4e, f), which clearly identifies a 31 nm thick SEI consisting of an 8 nm amorphous organic layer and a 23 nm inorganic layer composed mainly of ZnS, ZnO, and Zn3N2. The inner inorganic layer serves as a robust barrier against dendritic intrusion, while the outer organic layer accommodates mechanical stress and buffers volume fluctuations during Zn plating/stripping. Structure and composition of SEI layer in ZSLM electrolyte. a 3D variation of the strength of TOF-SIMS after 30 cycles of Zn negative electrode. XPS spectra of Zn negative electrode surface: b C 1 s, c S 2p, and d N 1 s. e TEM images of bilayer SEI. f HRTEM images of amorphous organic component, ZnO, ZnS, and Zn3N2 within the formed SEI. Source data are provided as a Source Data file. To gain deeper insight into the interfacial behavior and decomposition characteristics of L-Cys on the Zn surface, MD simulations were performed. According to the number density distributions derived from the MD model in Supplementary Figs. 43 and 44, L-Cys molecules exhibit a significant concentration peak at ~13 Å, indicating their strong interfacial affinity and preferential adsorption onto the Zn surface, thereby contributing directly to SEI formation. MP shows partial enrichment near the interface, while a significant depletion of H2O is observed in the same region, which suggests that the co-presence of L-Cys and MP effectively suppresses water adsorption, thereby reducing parasitic side reactions and facilitating the formation of a dense SEI.41 A deeper understanding of the interfacial behavior of L-Cys was obtained by calculating the adsorption energies of its three key functional groups (-SH, -COOH, and -NH2) on the Zn (002) surface, as shown in Supplementary Fig. 45a and Supplementary Data 78–80. The -NH2 group shows the highest adsorption energy (−1.26 eV), followed by -COOH (−1.15 eV) and -SH (−0.98 eV), indicating that L-Cys exhibits strong and multi-site surface affinity. Notably, the -NH2 group has the strongest adsorption energy and preferentially anchors on the Zn surface, thus promoting the formation of Zn3N2. In contrast, the -COOH and -SH groups subsequently participate in the formation of ZnO and ZnS due to their lower surface affinity. Moreover, the ESP distribution (Supplementary Fig. 45b) highlights highly electronegative regions around the -NH2, -COOH, and -SH groups, suggesting that these sites are prone to interacting with the Zn surface and initiating reduction reactions. To directly observe this process at the atomic level, MD simulations were conducted to study the formation of an inorganic-rich SEI on the Zn negative electrode in the ZSLM electrolyte. As shown in Fig. 5a and Supplementary Fig. 46, the L-Cys molecule (circled in white) adsorbed on the Zn surface exhibits elongated C-N and C-S bonds after 100 fs, indicating the beginning of decomposition. By 300 fs, N and S atoms are found to be coordinated with surface Zn atoms, resulting in the formation of Zn3N2 and ZnS. This evolution is further supported by the interatomic distance profiles in Supplementary Fig. 47, which track the dissociation of C-N and C-S bonds and the formation of Zn-N and Zn-S bonds, confirming the formation of an inorganic-rich interfacial SEI layer.42 Formation and function of SEI in ZSLM electrolyte. a Various snapshots of the AIMD simulations on the interface between ZSLM and Zn negative electrodes to illustrate the reaction process of forming the inorganic-rich SEI layers. b ELF distributions for different SEI compositions. c Zn2+ diffusion energy barrier related to different SEI compositions. d Schematic illustration of the Zn2+ solvation structure and interfacial behavior in conventional ZS electrolyte versus ZSLM electrolyte. Source data are provided as a Source Data file. FES was conducted to reveal the impact of inorganic SEI composition on Zn2+ concentration distribution and interfacial ion transport (Supplementary Fig. 48).43 In the ZS electrolyte, the Zn(OH)2-rich SEI exhibits irregular and inhomogeneous Zn2+ distribution, which may lead to localized charge accumulation and promote dendrite growth. In contrast, ZnO, ZnS, and especially Zn3N2, formed in the ZSLM electrolyte, support a more uniform Zn2+ concentration gradient, indicating stabilized interfacial ion transport. Notably, Zn3N2, which preferentially forms at the early stage of SEI development, contributes to a flatter and more stable Zn2+ concentration field, mitigating local concentration fluctuations and reducing dendrite formation. In addition to improving Zn2+ distribution, these inorganic species also serve as a structurally robust foundation for subsequent SEI evolution. The formation of an inner inorganic layer composed of Zn3N2 and ZnS facilitates the subsequent deposition of a stable outer organic-rich layer, resulting in a robust bilayer SEI structure, which physically shields the Zn surface from water and parasitic attack, effectively suppressing HER.44 The electronic environments of these SEI components were further investigated by examining their electron localization function (ELF) distributions (Fig. 5b). The Zn(OH)2 layer shows strong electron localization around oxygen atoms, hindering Zn2+ migration across the SEI. ZnO exhibits moderately delocalized electron density, but still retains significant localized bonding, which limits the continuity of ion transport pathways. In contrast, Zn3N2 presents a more delocalized and continuous ELF distribution, favoring Zn2+ diffusion. This trend is consistent with the calculated Zn2+ diffusion energy barriers across the different SEI compositions (Fig. 5c and Supplementary Data 81–84). Zn(OH)2 presents the highest barrier (1.18 eV), while Zn3N2 shows the lowest barrier (0.49 eV), confirming that Zn3N2 provides a more efficient Zn2+ conduction channel within the SEI film. Additionally, charge density difference analyses (Supplementary Fig. 49) were performed to examine the interfacial electronic interactions at the Zn/SEI interface. The Zn3N2-Zn interface exhibits a smooth and continuous charge redistribution profile, indicative of stable interfacial bonding and minimal electron scattering. In contrast, the Zn(OH)2-Zn interface shows fragmented and oscillatory charge separation, which reflects poor interfacial stability and a higher tendency toward parasitic reactions. Therefore, MP and L-Cys were selected through rational screening for their complementary roles in modulating Zn2+ solvation and SEI formation. Figure 5d provides a schematic illustration of this synergistic dual-additive mechanism, highlighting the coordinated regulation of interfacial chemistry and solvation structure. MP acts as a solvation regulator, where its polar C = O group interacts with water molecules to reconstruct the H-bond network and weaken Zn2+-H2O coordination. Simultaneously, L-Cys functions as an interfacial stabilizer, featuring multiple electron-rich functional groups (-SH-, -COOH, and -NH2) that strongly adsorb onto the Zn surface and undergo preferential reduction to initiate the formation of an inorganic-rich SEI. Specifically, the co-introduction of MP and L-Cys reconstructs the solvation shell into a [Zn(H2O)4-MP-L-Cys]2+ complex, reducing the hydration strength of Zn2+, thereby lowering the desolvation energy and accelerating ion transfer at the electrode interface. Meanwhile, L-Cys decomposes preferentially to generate an inner SEI layer rich in Zn3N2, ZnS, and ZnO, offering robust mechanical stability and fast Zn2+ transport within the SEI. By contrast, in ZS electrolyte, Zn2+ predominantly exists in a highly hydrated [Zn(H2O)6]2+ configuration, stabilized by an extensive H-bond network, which leads to a high desolvation energy barrier, poor Zn2+ interfacial transport, and heterogeneous deposition.45 Moreover, side reactions at the Zn surface promote the formation of a disordered SEI layer rich in Zn(OH)2, ZnO, and amorphous ZnSxOy, which lacks ionic conductivity and structural stability.37,46 Electrochemical performance of Zn negative electrodes The electrochemical reversibility of Zn||Cu cells was evaluated at 1.0 mA cm−2 and 1.0 mAh cm−2 using ZS and ZSLM electrolytes. As shown in Fig. 6a, b, the ZSLM electrolyte delivers a high initial CE (ICE) of 93.23% and maintains an average CE of 99.65% over 1200 cycles with low and stable polarization, indicating that the co-addition of L-Cys and MP effectively stabilizes Zn plating/stripping. In contrast, the ZS cell exhibits a lower ICE of 84.25%, along with increasing polarization and early failure after only 103 cycles (Supplementary Fig. 50), mainly due to uncontrolled dendrite growth and aggravated interfacial resistance. To further examine the effect of the dual-additive on Zn deposition behavior, the surface morphology and phase composition of Cu and Zn electrodes after different cycles in ZS and ZSLM electrolytes were investigated. SEM images of Cu positive electrodes after Zn deposition (Supplementary Fig. 51a, c) show that Zn deposits formed in the ZS electrolyte are irregular, porous, and non-uniform, with clear evidence of local dendritic growth and corrosion-induced degradation. In contrast, the ZSLM electrolyte yields smoother and more compact Zn layers, indicating more uniform Zn nucleation and better deposition reversibility (Supplementary Fig. 51b, d). Besides, the Zn foil similarly displays a denser and smoother surface morphology without detached protrusions after 100 cycles in the ZSLM electrolyte, compared to those in the ZS electrolyte, reflecting that the dual-additive can regulate Zn deposition behavior and suppress dendrite formation (Supplementary Fig. 51e–h). XRD analysis further demonstrates the formation of Zn4SO4(OH)6·5H2O by-products on Cu electrodes in the ZS electrolyte, which are markedly suppressed when MP and L-Cys are introduced (Supplementary Fig. 51i, j). Notably, the Cu electrode in the ZSLM electrolyte exhibits distinct diffraction peaks corresponding to the CuZn5 alloy phase, which offers homogeneous nucleation sites for compact Zn growth. On the Zn negative electrode, the increased intensity of the Zn (002) diffraction peak after cycling indicates a preferred orientation along the (002) plane, contributing to dense and oriented deposition.29 Electrochemical performances. a Voltage profiles of the Zn||Cu cell in ZSLM electrolyte with different cycles at 1 mA cm−2 and 1 mAh cm−2. b CE of the Zn||Cu cells using different electrolytes at 1 mA cm−2 and 1 mAh cm−2. Galvanostatic cycling of the Zn||Zn cells in different electrolytes at c 2 mA cm−2 and 2 mAh cm−2 and d 5 mA cm−2 and 5 mAh cm−2. e Rate performance of the Zn||Zn cells assembled with different electrolytes. f Comparison of cyclic reversibility in the recent reports. The sources of the literature data shown in this figure can be found in Supplementary Table 2. Source data are provided as a Source Data file. Building on the above results, the Zn||Zn symmetric cells were further tested to evaluate long-term electrochemical stability with increasing current density. At 2 mA cm−2 and 2 mAh cm−2 (Fig. 6c), the ZSLM electrolyte delivered stable Zn plating/stripping over 3000 h, with consistently low voltage polarization, demonstrating stable reversibility and interfacial compatibility. On the contrary, in the ZS electrolyte, severe voltage fluctuations and unstable interfacial behavior result in premature short-circuiting, reflecting uncontrolled Zn nucleation and dendritic growth in the absence of interfacial regulation. These results highlight the synergistic interaction between L-Cys and MP in simultaneously regulating Zn2+ solvation and interfacial chemistry, facilitating the formation of a robust SEI, which effectively suppresses parasitic reactions and promotes uniform Zn deposition. This performance advantage becomes more significant at 5 mA cm−2 and 5 mAh cm−2 (Fig. 6d), where the ZSLM electrolyte enabled stable Zn plating/stripping for over 600 h with minor voltage fluctuation. In contrast, the ZS electrolyte exhibited continuous polarization growth and eventual short-circuit failure. At high current density of 10 mA cm−2 and 10 mAh cm−2 (Supplementary Fig. 52), the ZSLM electrolyte still maintained a stable voltage profile for over 500 h, suggesting that the dual-additive modulates Zn nucleation, reduces the tendency for dendrite formation, and enhances Zn plating/stripping reversibility. The addition of L-Cys and MP significantly suppresses polarization, as shown in Fig. 6e, thereby greatly enhancing rate performance. The Zn||Zn symmetric cell in the ZSLM electrolyte exhibits stable Zn plating/stripping across various current densities. Notably, the cells in the ZSLM electrolyte demonstrate robust long-term stability of Zn plating/stripping across a wide range of current densities and areal capacities, outperforming the majority of previously reported electrolyte additives in Zn||Zn cells (Fig. 6f and Supplementary Table 2). The enhanced electrochemical stability is attributed to the synergistic interplay between L-Cys and MP, which governs Zn2+ solvation and interfacial chemistry, mitigating dendrite formation and parasitic side reactions. Well-regulated Zn nucleation and deposition dynamics contribute to homogeneous Zn plating/stripping behaviors. Electrochemical performance of Zn||I2 full cells To evaluate the practical applicability of the proposed electrolyte system, Zn||I2 full cells were assembled by coupling a Zn negative electrode with an I2 positive electrode, in which I2 is uniformly dispersed within the activated carbon (AC) matrix (Supplementary Fig. 53).47 From cyclic voltammetry (CV) curves (Fig. 7a), the Zn||I2 full cell using the ZSLM electrolyte exhibits sharp and symmetric redox peaks at 1.225 and 1.214 V, with a narrow peak-to-peak separation of 11 mV. In contrast, the broader and less intense peaks observed for the ZS electrolyte (1.237 V/1.201 V, 36 mV) reflect sluggish redox kinetics. The reduced voltage gap and higher peak current density in the dual-additive system indicate significantly lower electrochemical polarization and enhanced I−/I2 redox kinetics. Notably, as the scan rate increased from 0.1 to 2.0 mV s−1 (Supplementary Fig. 54), no additional redox peaks emerged, confirming that the introduction of L-Cys and MP does not alter the intrinsic redox behavior of the I2 positive electrode. The enhanced electrochemical kinetics are further evidenced by rate performance tests (Fig. 7b and Supplementary Fig. 55). The full cell using ZSLM electrolyte delivers higher capacities than that using ZS electrolyte, highlighting the enhanced rate performance and reversibility of the dual-additive system, especially under high current densities, which is consistent with the fast reaction kinetics observed in the Zn||Zn symmetric cells. Moreover, the ZSLM electrolyte enables the Zn||I2 full cell to deliver a reversible capacity of 229.1 mAh g−1 after 1000 cycles at 0.5 A g−1 (Fig. 7c, d) and to maintain stable capacity (166.5 mAh g−1) with high CE over 5000 cycles at 2.0 A g−1 (Fig. 7e and Supplementary Fig. 56a). In contrast, the full cell in ZS electrolyte experiences rapid performance degradation under the same conditions (Supplementary Fig. 56b and Supplementary Fig. 57). The corresponding ICE comparison at 0.5 and 2 A g−1 (Supplementary Fig. 58) further confirms the well-maintained voltage stability and electrochemical reversibility of the ZSLM electrolyte. Besides, the self-discharge performance of the Zn||I2 full cells in ZS and ZSLM electrolytes was assessed by charging to 1.4 V and resting for 24, 48, and 72 h. As shown in Supplementary Fig. 59, the capacity retention in ZS electrolyte decreased to 95.8%, 92.5%, and 89.3% after 24, 48, and 72 h of rest, respectively. In contrast, the capacity retention in ZSLM electrolyte remained high, reaching 99.2%, 98.1%, and 97.8%, indicating reduced parasitic reactions and improved long-term stability in the Zn||I2 full cell. Electrochemical performances of the Zn||I2 full cells. a CV curves and b rate performance of ZS and ZSLM electrolytes. c Cycling performance and CE at 0.5 A g−1 with different electrolytes and d corresponding galvanostatic charge-discharge (GCD) curves. e Cycling performance and CE at 2.0 A g−1 of ZS and ZSLM electrolytes. f GCD curves of Zn||I2 pouch cell at 2.0 A g−1 in ZSLM electrolyte. g Cycling performance and CE at 2.0 A g−1 for the Zn||I2 pouch cell in ZSLM electrolyte (insets: digital image of the pouch cells under harsh test conditions. Source data are provided as a Source Data file. In addition to the coin cell evaluation, the Zn||I2 pouch cell was evaluated to expand the evaluation of the application potential of ZSLM electrolyte (Supplementary Fig. 60). As shown in Fig. 7f, g, the pouch cell maintains stable cycling performance at 2.0 A g−1, retaining 93.1% of its capacity after 800 cycles relative to the first cycle. To further validate the practical applicability and safety of the proposed electrolyte, the Zn||I2 pouch cell was subjected to mechanical disturbance tests. As illustrated in Fig. 7g, the pouch cell in ZSLM electrolyte remained fully operational while powering a light-emitting diode display during physical disturbances, including needling and bending. Furthermore, video evidence (Supplementary Movie 1) confirms that the pouch cell in ZSLM electrolyte continues to function even after cutting, thereby demonstrating its robust electrochemical and mechanical stability. Overall, these results highlight the robust full cell performance, long-term stability, and mechanical durability of the ZSLM electrolyte, emphasizing its promise for integration into flexible and wearable Zn-based energy storage devices operating under demanding conditions. Discussion In summary, this study presented a molecularly engineered electrolyte strategy that integrates L-Cys and MP as synergistic additives to simultaneously modulate Zn2+ solvation structure and interfacial chemistry in AZMBs. L-Cys preferentially adsorbs onto the Zn surface and facilitates the formation of an inorganic-rich inner SEI layer (Zn3N2/ZnS/ZnO), while MP reorganizes the H-bond network to reduce water activity and promote Zn2+ desolvation. Combined experimental and theoretical analyses, including MD, DFT, XPS, TOF-SIMS, and FES, confirmed that the designed electrolyte leads to improved Zn2+ migration pathways and enhanced interfacial stability. Consequently, the Zn||Zn symmetric cell exhibited stable Zn plating/stripping (>3000 h at 2 mA cm−2 and 2 mAh cm−2), and the Zn||I2 pouch cell achieved robust durability and mechanical resilience. This work offers a generalizable approach for rational electrolyte design in aqueous multivalent metal batteries, bridging solvation thermodynamics with SEI engineering to achieve long-term electrochemical stability. Methods Characterizations The spectroscopic information of different solutions was demonstrated using a FTIR (Thermo Scientific iN10) and a liquid-state NMR (BRUKER AVANCE 400). XRD patterns were obtained by using a powder X-ray diffractometer (Bruker-D8 ADVANCE), covering a scanning range of 5° to 80°. Operando visualization of Zn deposition behavior was carried out by using an optical microscope (Nikon, ECLIPSE LV 150 N). The surface profiles of the Zn electrodes after cycling were detected by using a scanning probe microscope (Veeco D3100). In-situ Raman measurements were conducted using a RENISHAW INVIA1 123-11 on transparent Zn||Cu cells, with the laser focused near the Cu/electrolyte interface and signals collected by three-scan integration during discharge (5 mA cm−2 and 5 mA hcm−2). 3D and 2D surface mapping and depth-resolved elemental analysis of the Zn negative electrode were conducted using a time of flight secondary ion mass spectrometry (TOF-SIMS 5 (ION-TOF GmbH, Germany)). A Bi3+ liquid metal ion gun was used for imaging, and a 1 kV Cs+ sputtering source enabled depth profiling over a 100 μm × 100 μm area for up to 900 s. Measurements were performed in negative ion mode to track the distribution of organic (CH− and CN−) and inorganic (ZnOH−, ZnN−, ZnS−, and ZnO−) SEI species. Surface chemistry was analyzed by XPS (Thermo ESCALAB 250 XI). To investigate the internal composition of the SEI, the samples were sputtered with Ar+ for 40, 80, and 120 s. TEM (FEI Tecnai G2 F30) was used to investigate the microstructures of bilayer SEI on the Zn negative electrode in ZSLM. Viscosity measurements were performed using a Thermo Scientific HAAKE MARS 60 rheometer with Peltier temperature control and a 27 mm double-gap coaxial cylinder. CLSM images of the Zn electrodes after Zn plating/stripping in ZS and ZSLM electrolytes were obtained using an Olympus LEXT OLS5000 microscope. Preparation of electrolytes Amounts of L-Cys (98%, Sigma-Aldrich) and MP (99%, Sigma-Aldrich) were dissolved in 2 M ZS (99%, Sigma-Aldrich) under constant magnetic stirring at 25 ± 2 °C until the solution became clear and homogeneous in an open atmosphere. The dual-additive electrolyte (ZSLM) was formulated with a composition of ZS, 50 mM L-Cys, and 100 mM MP. To elucidate the individual effects of each additive, the following control electrolytes were also prepared under identical conditions: ZS, ZS + 50 mM L-Cys (ZSL), ZS + 100 mM MP (ZSM), ZS + 50 mM Asp (98%, Sigma-Aldrich) (ZSA), ZS + 50 mM Ser (98%, Sigma-Aldrich) (ZSS), ZS + 100 mM PC (99%, Sigma-Aldrich) (ZSP), and ZS + 100 mM DME (99.5%, Sigma-Aldrich) (ZSD). All the electrolyte formulations were stirred for at least 2 h to ensure complete dissolution and equilibration prior to electrochemical measurements. Preparation of AC/iodine (ACI) positive electrode For the positive electrode material, AC (98%, Sigma-Aldrich) was mixed and ground with iodine (99.8%, Sigma-Aldrich) powder to obtain a fine mixture. The mixture was then sealed in an ampoule. ACI was collected after drying at 80 °C for 8 h, and the weight ratio of iodine in ACI was ~30%. A mixture of ACI, conductive carbon (Super P, Canrd Technology Co. Ltd.) and polyvinylidene fluoride (Canrd Technology Co. Ltd.) in an 8:1:1 mass ratio in a N-methyl-2-pyrrolidone (NMP, 99.5%, Sigma-Aldrich) solution was continuously magnetically stirred for 24 h before being doctor blade coated onto graphite paper (99%, 50 μm thickness, Canrd Technology Co. Ltd.), which did not adsorb iodine, and then vacuum-dried at 60 °C for 8 h to prepare the positive electrode. Assembly of cells Electrolyte solutions were prepared one day prior to cell assembly and stored in the dry cabinet at 25 °C. Zn foils were mechanically polished and rinsed in ethanol before use. Symmetric and asymmetric cell configurations were assembled using 2032-type coin cells, which inherently include a stainless steel spacer (1.0 mm thickness), a spring-loaded plunger (stainless steel, 1.5 kN constant force), the working electrodes, and a glass fiber filter (Whatman GF/D, Canrd Technology Co. Ltd) separator with a diameter of 16 mm. For the Zn||Zn symmetric cell, two pieces of Zn foils (99.9%, Canrd Technology Co. Ltd.) with a thickness of 100 μm and a diameter of 12 mm were used as the two electrodes. For the Zn||Cu asymmetric cell, one Zn foil was replaced by a Cu foil (20 μm thickness and 12 mm diameter, Canrd Technology Co. Ltd.). The volume of the electrolyte was fixed at 80 μL. Zn||I2 full cells were assembled using I2 as the positive electrode, Zn foil as the negative electrode, and a glass fiber filter as the separator. The mass loading of I2 in coin cells was ~3.0–3.5 mg cm2. The single-layer Zn||I2 pouch cells were assembled by stacking one I2 positive electrode (6 × 6 cm, loading of ~15.8 mg cm−2) and one polished Zn negative electrode (6 × 6 cm, 20 μm) on either side of a glass fiber filter separator (6.2 × 6.2 cm). The thickness of the separator after compression was ~500 μm, with an average pore size of 1.0 μm. The assembled stacks were placed in an Al-plastic bag and filled with electrolyte at a dosage of 20 μL mAh−1. After degassing and heat sealing, the pouch cells were allowed to rest for 2 h to ensure complete wetting. Galvanostatic cycling stability tests were then conducted under a uniform pressure of about 2 kPa to ensure satisfactory contact among the cell components. Electrochemical measurements Assembled coin cells (CR2032) for the Zn||Zn symmetric cells, Zn||Cu half cells, and Zn||I2 full cells were prepared in an open-air atmosphere. All the cycling data were independently repeated at least three times using electrolytes prepared from two separate batches. The curves shown in the main figures correspond to representative cells with performance close to the average values. For the electrochemical performance, the charge/discharge, CE, and cycling performance of the cells were measured on WonaTech Automatic cell test instrument (WBCS3000). The Zn||I2 cells were cycled at 0.6 and 1.4 V. The CV, LSV (sweep rate of 5 mV s−1), CA (at overpotentials of −150 mV) and EIS were performed on an Ivium Stat electrochemical workstation. EIS measurements were performed on Zn||Zn symmetric cells at different temperatures in a frequency range from 0.1 to 100 kHz with five data points per decade and an amplitude of 5 mV under stable open-circuit voltage conditions. For some tests (Tafel and LSV), a three-electrode system was employed using Zn foil as the working electrode, Pt wire as the counter electrode, and Ag/AgCl as the reference electrode. The CV was performed using Zn||I2 cells under various scan rates (0.1–2.0 mV s−1). All the measurements were performed at 25 ± 2 °C. DFT calculation DFT simulations were performed using DMol3 module of BIOVIA Material Studio.48,49 Generalized gradient approximation of Perdew-Burke-Ernzerhof was used to describe the exchange–correlation energy function.50–52 The nuclear processing method uses all-electron relativistic effects.53 The Basis set was selected DNP with 3.5 Basis file. All electrons were used to treat the core electrons. First, we carried out geometric optimization for various organic and amino acid molecules, and calculated their ESP, HOMO and LUMO to predict the reactivity and stability of the molecules in chemical reactions. The energy change, maximum force, and displacement convergence tolerance for geometry optimization were set at 1.0 × 10-5 Ha, 0.002 Ha Å−1, and 0.005 Å, respectively. The electronic self-consistent field convergence threshold was set to 10-6 hartree. To speed up the convergence, thermal smearing was applied to the orbital occupation with a smearing parameter of 0.005 hartree. Moreover, we calculated the binding energy (Eb) between all the organic and amino acid molecules with Zn2+ and H2O as follows: Additionally, to describe the anchoring behavior of various molecules on the surface of Zn (002), 4 × 4 × 1 supercells were constructed from Zn (002) crystal surface and various Zn compounds. A vacuum layer of 15 Å was also set to avoid the influence of periodic structures. And the k-points separation for Zn (002) and other Zn compounds was 2 × 2 × 1. The energy, force, and displacement convergence tolerance for geometry optimization were set at 1.0 × 10−5 Ha, 0.002 Ha Å−1, and 0.005 Å, respectively. In the process of geometric optimization, all the structures are completely relaxed. To describe the anchoring behavior of various molecules on the surface of Zn (002), the adsorption energy is defined as follow Eq. (2): where Ecomplex, EZn(002), and EMolecule are the energies of the absorption complex structures, Zn (002), and different molecules, respectively. MD simulations Models of electrolyte solutions and the adsorption process of L-Cys and MP molecules on the Zn (002) surface were investigated by MD. And, all the MD simulations were performed using Forcite module of BIOVIA Material Studio. The composite structures were built by Amorphous Cell module. Firstly, according to the coordinated water of Zn salt, a box with a size of 74.6 × 74.6 × 74.6 Å3 was constructed. It contains 400 Zn2+, 400 SO42−, and 5000 H2O molecules. The COMPASS III force field was utilized to describe the interaction between ZnSO4, L-Cys, and MP molecules.54,55 The van der Waals interactions were controlled with an atom-based cutoff distance of 15.5 Å, and the electrostatic interactions were considered using the Ewald method. The cutoff length was set to 1.0 nm. And, the integration time step was set to 1 fs. The convergence tolerances for the energy and maximum force were set as 0.0001 kcal mol−1 and 0.005 kcal mol−1 Å−1, respectively. All the systems were annealed from 0 to 298 K over a period of 0.5 ns and subsequently equilibrated for an additional 0.5–1 ns to achieve stable equilibrium conditions (The isothermal–isobaric ensemble simulation was used to equilibrate the temperature and pressure). The canonical ensemble (NVT) was simulated for 1 ns using the Berendsen thermostat, with a time step of 1 fs at 298 K. The van der Waals interactions were estimated using atom-based summation with a cutoff distance of 15.5 Å, whereas the particle–particle particle–mesh approach with a 0.001 kcal mol−1 accuracy was utilized for electrostatic interactions. After the NVT simulation, the RDF (g(r)) and coordination number (n(r)) were calculated using a track file at intervals of 1 fs. The coordination number of molecules was calculated according to Eq. (3): where ρ(r) is the local density of atoms. Computational simulations For FESs, the “tertiary current distribution” physics interface in COMSOL Multiphysics 6.0 was utilized to model both the current density field and ion concentration field. The potential and concentration distributions were governed by the Nernst-Planck and Poisson equations,56 as described below: where Ji is the ion flux, D is the diffusion coefficient, ci is the Zn2+ concentration, R is the ideal gas constant, T is the temperature, F is the Faraday constant, ϕ is the electrolyte potential, and σ is the interfacial mobility. All the models consider the simplified bare Zn electrode surface structure, with four protrusions set on the electrode surface. The electrolyte served as the main body in the simulation environment. The initial concentration of Zn2+ was set to 2.0 M. The exchange current density across the cell was set to 2 mA cm−2, while the system temperature was maintained at 273 K. Supplementary information Supplementary Information Description of Additional Supplementary Files Supplementary Data Supplementary Movie 1 Transparent Peer Review file Source data Source Data Acknowledgements The authors would like to thank the support from the National Research Foundation of Korea (NRF) grant funded by the Korean Government (MSIT) (no. 2018R1A6A1A03025708), received by J.-S.Y., the faculty startup funds from the Yangzhou University, received by J.-B.L., the National Natural Science Foundation of China (52502369), received by J.-B.L., and the Lvyangjinfeng Talent Program of Yangzhou, received by J.-B.L. Author contributions S.T.: conceptualization, methodology, data curation, and writing-original draft. J.L.: funding acquisition, conceptualization, and writing-review & editing. T.W.: conceptualization and visualization. T.Y.W. and H.M.P.: conceptualization and methodology. W.X. and Y.X.: data curation and formal analysis. S.X.: investigation and conceptualization. J.S.Y.: funding acquisition, supervision, conceptualization, and writing-review & editing. H.P.: conceptualization, writing-review & editing, and validation. Peer review Peer review information Nature Communications thanks Junchao Zheng and the other, anonymous reviewers for their contribution to the peer review of this work. [A peer review file is available]. Data availability All data generated in this study are provided in the manuscript, Supplementary Information, and Source Data file. Source data are provided with this paper. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Jiabao Li, Email: jiabaoli@yzu.edu.cn. Jae Su Yu, Email: jsyu@khu.ac.kr. Huan Pang, Email: panghuan@yzu.edu.cn. Supplementary information The online version contains supplementary material available at 10.1038/s41467-026-73025-z. References Associated Data Supplementary Materials Supplementary Information Description of Additional Supplementary Files Supplementary Data Supplementary Movie 1 Transparent Peer Review file Source Data Data Availability Statement All data generated in this study are provided in the manuscript, Supplementary Information, and Source Data file. Source data are provided with this paper.