Zinc-oligoether carboxylate salts as electrolyte additives for aqueous Zn metal batteries Authors: John Brown, Ivette Aguilar, Juan Forero-Saboya, Chhail Bihari Soni, Hugo Chotard, Yuhuan Wang, Everett Zuras, Jean-Marc Noël, Jean-Marie Tarascon, Alexis Grimaud Published: 2025 ABSTRACT Aqueous zinc-metal batteries, which pair a water-based electrolyte with an earth-abundant metal, offer a sustainable energy storage platform. However, hydrogen evolution and dendrite growth during zinc plating limit stability and cycle life. Achieving durable performance requires a comprehensive approach that promotes forming a robust solid electrolyte interphase layer. Here, we introduce zinc oligoether carboxylates as tuneable dual-function electrolyte additives. The carboxylate group promotes the formation of a zinc sulphide-based interphase through the controlled decomposition of sulphate anions, while the oligoether chain suppresses water activity at the electrode surface, as revealed by spectroscopic and electrochemical microscopy analyses. By varying the chain length, we find that the longest compound, zinc (2-[2-[2-(2-methoxyethoxy)ethoxy]ethoxy]acetate)2, extends the lifetime of symmetric zinc cells from 200 to 1000 hours at a current density of 0.624 mA.cm-2 and a capacity of 2.5 mAh.cm-2. Furthermore, zinc–copper cells achieve 99.6% efficiency at 5 mA.cm-2 and 10 mAh.cm-2. This improvement also benefits zinc–manganese dioxide, extending from 30 to 200 cycles. These results demonstrate a molecular strategy for stabilising zinc interfaces, hence advancing practical aqueous zinc–manganese dioxide batteries. Introduction Aqueous rechargeable zinc metal batteries (AZMBs) are promising candidates for large-scale stationary energy storage, with metallic Zn metal electrode offering high gravimetric capacity (820 mAh.g−1), volumetric capacity (5855 mAh.cm−3), and a low electrode potential (−0.76 V vs. SHE). Combining a water-based electrolyte with an earth-abundant element further positions AZMBs as a viable green and sustainable battery technology1,2. However, widespread adoption, either as Zn||MnO2 or Zn-air, is hindered by significant challenges, including limited cycling lifespan and reduced coulombic efficiency (CE%), which are largely attributed to the electrochemical behaviour of the zinc metal electrode. At the Zn–electrolyte interface, two major challenges hinder the performance, dendrite growth and corrosion. The first challenge lies in regulating the kinetic parameters to favour the formation of preferential Zn facets, thereby optimising the nucleation process. It is well known that zinc nucleation requires high energy, which promotes uneven deposition and dendritic growth, particularly at low current densities and leading many studies to employ high current densities when testing Zn cells3–6. However, such conditions do not reflect real-world applications. For example, commercial stationary systems—a potential application for AZMBs—typically operate at low current densities7. The second key issue is the inherently low thermodynamic stability of Zn metal electrodes, which induces corrosion regardless of pH. In acidic media, zinc corrosion leads to parasitic hydrogen evolution reaction (HER) and the formation of a passivating zinc hydroxide sulfate layer (Zn4(OH)6(SO4).xH2O–ZHS) on the surface8. While the ZHS is essentially a solid electrolyte interphase (SEI) that does passivate the surface, its effectiveness as a protective layer is limited. As a solid acid–base species, it is highly sensitive to pH fluctuations at the interface, which compromises its stability and the overall performance of Zn-based batteries9,10. Therefore, strategies to improve or replace this native SEI are necessary for enhanced long-term performance11. A multi-faceted approach is required to comprehensively solve the complexities at the dynamic Zn-electrolyte interface, involving both electrolyte design and Zn surface engineering. In terms of electrolyte design, significant efforts must be made to control the pH and promote the formation of Zn complexes, in order to regulate cation–water and water–water interactions. With this in mind, glymes and polymers—due to their dual hydrophobic/hydrophilic nature—are known to influence water structuring, reducing the accumulation of water molecules and thereby limiting water splitting12–16. Nevertheless, this approach cannot fully prevent the HER, which remains predominant in all aqueous systems, especially at low current densities or during rest periods, even when using water-in-salt electrolytes17. AZMBs thus necessitate the formation of a SEI, that is uniform, compact, and electronically insulating yet ionically conductive in order to prevent persistent HER at negative electrode surface9,18. In zinc sulphate electrolytes, the most effective SEI demonstrated so far is ZnS, typically formed via super-acid additives. For example, Nazar et al. achieved ZnS formation by adding 0.05–0.1 M N,N-dimethylformamidium trifluoromethanesulfonate, while Ren et al. employed 0.1 mol.kg−1 HTFSI. These additives introduce super-acidity to locally generate protons on the Zn surface, facilitating SO42− decomposition to form ZnS. Simultaneously, their hydrophobic –CF3 groups repel water molecules. At first glance, increasing the acidity of the electrolyte may seem counterintuitive, as it would enhance the corrosion of zinc rather than reduce it. However, the acidity of the additive can lead to fast and efficient SEI formation, which results in long-term cell stability19,20. Here, we draw inspiration from ZnS-based SEI strategies to present a robust protective layer formed simply through sulphate ion decomposition. This approach offers the advantage of requiring no major modifications to the widely used 2 M ZnSO4 electrolyte, while yielding an SEI that is largely independent of pH variations. We introduce a family of compounds that simultaneously feature a carboxylic acid group and a glyme chain. These compounds, referred to as zinc oligoether carboxylate salts, have been previously studied as ionic liquids with Li and Na analogues, but their potential as additives to prevent dendrite formation in AZMBs has not yet been explored21,22. The carboxylic acid (COOH) group (pKa ≈ 4–5) can buffer the acid/base dynamics at the Zn surface and mitigate the formation of passivating ZHS layer, while facilitating the effective formation of a Zn-based SEI23–26. In parallel, we hypothesise that the oligoether chain helps to exclude water molecules from the Zn interface, thereby reducing corrosion, suppressing dendrite growth, and promoting uniform Zn electrodeposition. Finally, oligoether carboxylates are fluorine-free, aligning with the green and clean ethos of AZMBs. Discussions The synthesis of the zinc oligoether carboxylate salts, fully detailed in the “Methods”, was adapted from the synthesis of alkali-metal analogues and is solvent-free and scalable22. Briefly, 2-(2-(2-methoxyethoxy)ethoxy)ethanol (n = 3) is reacted with sodium metal under an inert atmosphere, followed by the addition of chloroacetic acid, eventually yielding 2-[2-[2-(2-methoxyethoxy)ethoxy]ethoxy]acetic acid (HMEEA) after treatment with phosphoric acid. A cation exchange reaction is then used to form Zn(MEEA)2. The procedure (Fig. 1) is adaptable whereby different chain lengths can be synthesised by varying the length of the glyme precursor; for example n = 1, zinc (2-(2-methoxyethoxy)acetate)2 (Zn(MA)2) and n = 2, zinc (2-[2-(2-methoxyethoxy)ethoxy]acetate)2 (Zn(MEA)2), were also synthesised. Successful synthesis was confirmed through nuclear magnetic resonance (NMR) spectroscopy, mass spectrometry, and elemental analysis (Figs. S1–6). Differential scanning calorimetry (DSC) revealed melting and glass transition temperatures decreasing with longer chains. Zn(MEEA)2 appears to be an ionic liquid with glass transition temperatures below 100 °C, similarly to the Li and Na analogues previously reported21,22. Thermogravimetric analysis (TGA) demonstrated high thermal stability, with decomposition temperatures around 300 °C for all salts (Fig. S7).Fig. 1Synthesis procedure for zinc oligoether carboxylate salts.The electrochemical behaviour of these compounds as primary electrolyte salts was first examined by preparing 1 M Zn(MA)2, Zn(MEA)2, and Zn(MEEA)2 aqueous solutions. All three electrolytes exhibited low ionic conductivities: Zn(MA)2 = 8.0 mS.cm−1, Zn(MEA)2 = 6.1 mS.cm−1 at 25 °C, and Zn(MEEA)2 = 3.4 mS.cm−1 at 25 °C (Fig. S8a). While Zn||Zn cells could be cycled in these electrolytes, cell life was limited to a maximum of 100 h before shorts were observed (Fig. S8b). The exceptionally low ionic conductivity of these salts, particularly compared to 2 M ZnSO4 (48 mS.cm−1 at 25 °C), likely leads to poor ion distribution during Zn electrodeposition, which is associated with dendrite formation48,49, excluding the use of zinc oligoether carboxylates as salts, but rather as additives. Synthesis procedure for zinc oligoether carboxylate salts. The electrochemical behaviour of these compounds as primary electrolyte salts was first examined by preparing 1 M Zn(MA)2, Zn(MEA)2, and Zn(MEEA)2 aqueous solutions. All three electrolytes exhibited low ionic conductivities: Zn(MA)2 = 8.0 mS.cm−1, Zn(MEA)2 = 6.1 mS.cm−1 at 25 °C, and Zn(MEEA)2 = 3.4 mS.cm−1 at 25 °C (Fig. S8a). While Zn||Zn cells could be cycled in these electrolytes, cell life was limited to a maximum of 100 h before shorts were observed (Fig. S8b). The exceptionally low ionic conductivity of these salts, particularly compared to 2 M ZnSO4 (48 mS.cm−1 at 25 °C), likely leads to poor ion distribution during Zn electrodeposition, which is associated with dendrite formation48,49, excluding the use of zinc oligoether carboxylates as salts, but rather as additives. 5 wt% of Zn(MA)2, Zn(MEA)2 and Zn(MEEA)2 was added to a base aqueous electrolyte of 2 M ZnSO4. Upon addition of the additives, a small drop in conductivity from 48 mS.cm−1 to ~40 mS.cm−1 at 25 °C is observed, likely related to the oligoether chains which cause a decrease in ionic mobility (Fig. S9). When testing Zn||Zn cells, both applied current density and areal capacity play critical roles in the onset of dendrite formation and short circuits. Current density, in particular, has been shown to significantly influence plating/stripping performance, with higher values increasing the nucleation overpotential and, consequently, the number of nucleation sites6,27. Additionally, the areal capacity, should be in the range of 2–3 mAh.cm−2 to match the capacity achieved by full cells under practical plating/stripping conditions7. Furthermore, to mitigate the HER, large current densities are often used to artificially exacerbate cycle life28,29. With this in mind, a low current density of 0.624 mA.cm−2 was chosen, combined with a high areal capacity of 2.5 mAh.cm−2. This converts to a C-rate of 0.25 C. Under these conditions, the control 2 M ZnSO4 electrolyte experiences a short circuit after 200 h of plating/stripping. In contrast, oligoether carboxylate additives enable a significant improvement in cycle life, with 1000 h achieved with no soft shorts detected (Fig. 2a, b).Fig. 2Zn stripping/plating performance in Zn||Zn and Zn||Cu.a Zn||Zn symmetric cells for 2 M ZnSO4 (green), 2 M ZnSO4 + 5 wt% Zn(MA)2 (dark green), 2 M ZnSO4 + 5 wt% Zn(MEA)2 (blue) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple) at 0.624 mA.cm−2/2.5 mAh.cm−2. b Plating/stripping curves at 190–210 h (top) and 900–920 h (bottom). c CE% determination with 2 M ZnSO4 + 5 wt% Zn(MA)2, 2 M ZnSO4 + 5 wt% Zn(MEA)2 and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 in Zn||Cu. d CE% for 2.5, 5 and 10 wt% of Zn(MEEA)2 data presented as mean from 3 cells with +/-SD. e Long-term plating/stripping of 5 wt% Zn(MEEA)2 in Zn||Cu at 5 mA.cm−2/10 mAh.cm−2 with CE% and f Voltage profiles 5 wt% Zn(MEEA)2 in Zn||Cu at cycles 1, 10, 100, and 250. All battery cycling took place under ambient conditions. Zn stripping/plating performance in Zn||Zn and Zn||Cu. a Zn||Zn symmetric cells for 2 M ZnSO4 (green), 2 M ZnSO4 + 5 wt% Zn(MA)2 (dark green), 2 M ZnSO4 + 5 wt% Zn(MEA)2 (blue) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple) at 0.624 mA.cm−2/2.5 mAh.cm−2. b Plating/stripping curves at 190–210 h (top) and 900–920 h (bottom). c CE% determination with 2 M ZnSO4 + 5 wt% Zn(MA)2, 2 M ZnSO4 + 5 wt% Zn(MEA)2 and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 in Zn||Cu. d CE% for 2.5, 5 and 10 wt% of Zn(MEEA)2 data presented as mean from 3 cells with +/-SD. e Long-term plating/stripping of 5 wt% Zn(MEEA)2 in Zn||Cu at 5 mA.cm−2/10 mAh.cm−2 with CE% and f Voltage profiles 5 wt% Zn(MEEA)2 in Zn||Cu at cycles 1, 10, 100, and 250. All battery cycling took place under ambient conditions. To accurately estimate the coulombic efficiency (CE%), a method described by Adams et al. and Xu et al. is employed (see “Methods”)30,31. Following this protocol, the base electrolyte (2 M ZnSO4) was unable to complete the protocol due to dendrite formation (Fig. S10), in line with previous studies30,32. Meanwhile, 5 wt% of Zn(MA)2, Zn(MEA)2 and Zn(MEEA)2 achieved 99.6%, 99.7% and 99.9% CE, respectively (Fig. 2c); the increased CE with chain length suggests that the latter has a consequence on the plating/stripping. To understand the role of the chain length, the nucleation of Zn2+ was studied using cyclic voltammetry (CV) on a glassy carbon electrode at a scan rate of 5 mV.s−1 (Fig. S11). As the chain length increases, there is an increase in the Zn2+ nucleation overpotential with 12, 19 and 25 mV for Zn(MA)2, Zn(MEA)2 and Zn(MEEA)2, respectively. This increase in overpotential could be attributed to the oligoether carboxylates interacting with the electrical double layer (EDL), particularly the inner Helmholtz plane (IHP). According to the Gouy-Chapman-Stern theory of the EDL, in 2 M ZnSO4, hydrated Zn2+ cations approach the negatively charged zinc metal surface through electrostatic interactions, occupying the outer Helmholtz plane while water molecules dominate the IHP, creating a thin EDL. This is reflected in the high EDL capacitance (EDLC) of 494 µF.cm2. The addition of 5 wt% of oligoether carboxylate reduces the EDLC to 277, 262 and 160 µF.cm2 for Zn(MA)2, Zn(MEA)2 and Zn(MEEA)2, respectively (Fig. S12). This suggests that the longer the chain of the oligoether, the thicker the EDL, potentially due to increased displacing of water at the IHP. Indeed, the strong Lewis acidity of Zn2+ in [Zn(H2O)6]2+ complexes enhance water activity and thus promotes the HER. In contrast, the hydrophobic domain formed near the Zn surface due to the oligoether carboxylate drives Zn2+ de-solvation, preventing [Zn(H2O)6]2+ from reaching the interface. As a result, HER is suppressed, minimising electron loss and thereby improving CE%. Having established that the highest CE% are achieved when using Zn(MEEA)2, the effect of additive’s concentration was probed; 5 wt% was determined to be the optimum, with 2.5 and 10 wt% giving a lower CE of 99.6 and 99.4%, respectively (Figs. 2d and S13). When testing this optimised electrolyte composition under more aggressive conditions, i.e., current density of 5 mA.cm−2 and an areal capacity of 10 mAh.cm−2, Zn||Cu cells were found to cycle for 250 cycles, totalling a cumulative capacity of 2500 mAh.cm−2 with an average CE of 99.6% (Fig. 2e), while the control 2 M ZnSO4 could only achieve 5 cycles before dendrites formation (Fig. S14). The voltage profiles validate the reversible Zn plating/stripping at such high current density and capacity in the presence of Zn(MEEA)2 (Figs. 2f and S15). This high cumulative capacity in tandem with high areal capacity, matches well with current literature examples and highlights the performance of Zn(MEEA)2 (Table S1). To ensure that both the oligoether chain and the carboxylate function play a role in the enhanced Zn plating/striping behaviour, results obtained with Zn(MEEA)2 additive were compared with those obtained with glymes, which share the flexible, long ether chains but lack the COO− group. For comparison, tetraethylene glycol dimethyl ether (TEGDME), which has the same ether chain length (n = 3) as Zn(MEEA)2, was selected. The stripping/platting of Zn||Zn cells containing 2 M ZnSO4 + 5 wt% TEGDME at 2.5 mA.cm−2/2.5 mAh.cm−2, only managed 75 h, a tenfold decrease compared to Zn(MEEA)2 (750 h) (Fig. S16a). This performance difference between the oligoether carboxylate and glyme is further confirmed when measuring the CE of glyme-based systems, for which 5 wt% and 10 wt% TEGDME showed a CE% of only 95.5% and 96%, respectively (Fig. S16b). Previous studies using TEGDME attributed performance gains to solvation effects, where TEGDME molecules displace water from Zn2+ solvation shell, thereby preventing the HER and enhancing Zn electrodeposition12,14. This effect is more pronounced at higher TEGDME concentrations, which supports the use of TEGDME as a co-solvent rather than an additive. Nevertheless, while the addition of TEGDME to 2 M ZnSO4 does improve Zn plating/stripping, the performance pales in comparison to Zn(MEEA)2. To test whether performance could be reproduced by combining separate molecules with oligoether and carboxylate functions, we evaluated 2 M ZnSO4 with 0.1 M TEGDME and 0.1 M Zn(Ac)2. The CE remained similar to using TEGDME as a lone additive (95.6%) (Fig. S17). These results highlight the necessity of a single molecule bearing both oligoether and carboxylate groups, as separate additives cannot reproduce the enhanced performance of Zn(MEEA)2. To understand the role played by the oligoether carboxylate additive, physio-chemical analysis was carried out. To probe the effect of Zn(MEEA)2 on solvation structure, Fourier-Transform Infrared Spectroscopy (FT-IR) was used, where no significant changes to water bands were observed (Fig. S18). Likewise, NMR showed no changes to the water structuring (Fig. S19). This suggests that, at this concentration, the oligoether carboxylate additive have a zero-to-minimal effect on the solvation structure. To understand the enhancement in Zn electrodeposition, Zn negative electrode were recovered after stripping/platting at 0.624 mA·cm−2/2.5 mAh·cm−2 for 20 cycles and the morphology was examined via scanning electron microscopy (SEM). The SEM images reveal mossy zinc growth and significant inhomogeneity on the negative electrode cycled with the control 2 M ZnSO4 electrolyte (Fig. 3a). In contrast, negative electrodes cycled with 5 wt% Zn(MEEA)2 display a uniform surface, characterised by a predominantly horizontal array of Zn crystals (Fig. 3b). This indicates anisotropic growth along the basal planes (corresponding to the 002 reflection) in the hexagonal close-packed lattice of zinc. X-ray diffraction analysis (Fig. 3c) reveals that the (002)/(100) Zn diffraction peak ratio decreased from 1.41 for the pristine to 1.28 in the control 2 M ZnSO4 electrolyte. Meanwhile, the ratio increased to 1.60 with 5 wt% Zn(MEEA)2, indicating that Zn(MEEA)2 promotes Zn growth with exposed (002) facets. While this difference is modest, we attribute it to the thick Zn metal (0.25 mm) used in the experiments, which allowed for facile disassembly and ensure the surface electrodes was easy to handle. Horizontal crystal growth along the Zn(002) plane has been tentatively linked to promoting chemical stability during Zn electrodeposition due to its lower surface energy compared to (100) facets33–35. To evaluate the effect of Zn(MEEA)2 on zinc nucleation and growth, chronoamperometry (CA) was conducted at −150 mV (Fig. 3d). In the ZnSO4 electrolyte, the current density increased continuously over 50 s before stabilising, indicating uncontrolled 2D diffusion and uneven dendrite growth caused by the “tip effect,” where Zn2+ preferentially deposits at low-energy sites, forming protrusions36–38. In contrast, adding 5 wt% Zn(MEEA)2 led to rapid nucleation and a stable current within 15 s, reducing 2D diffusion, which leads to a more uniform crystal growth, matching the earlier observations by SEM and XRD.Fig. 3Crystallographic analysis of the Zn negative electrodes.a SEM for Zn negative electrode cycled in 2 M ZnSO4 taken after 20 cycles at 0.624 mA.cm−2/2.5 mAh.cm−2 with picture of cycled Zn negative electrodes (plated electrode), b 2 M ZnSO4 + 5 wt% Zn(MEEA)2 taken after 20 cycles at 0.624 mA.cm−2/2.5 mAh.cm−2 with picture of cycled Zn negative electrodes (plated electrode) and c XRD patterns for pristine Zn negative electrode (grey), 2 M ZnSO4 (green) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple). d Chronoamperometry measurements with an applied potential of −150 mV for pristine 2 M ZnSO4 (green) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple). Crystallographic analysis of the Zn negative electrodes. a SEM for Zn negative electrode cycled in 2 M ZnSO4 taken after 20 cycles at 0.624 mA.cm−2/2.5 mAh.cm−2 with picture of cycled Zn negative electrodes (plated electrode), b 2 M ZnSO4 + 5 wt% Zn(MEEA)2 taken after 20 cycles at 0.624 mA.cm−2/2.5 mAh.cm−2 with picture of cycled Zn negative electrodes (plated electrode) and c XRD patterns for pristine Zn negative electrode (grey), 2 M ZnSO4 (green) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple). d Chronoamperometry measurements with an applied potential of −150 mV for pristine 2 M ZnSO4 (green) and 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple). The composition of Zn negative electrode surfaces after plating/stripping was then probed with X-ray photoelectron spectroscopy (XPS) and depth profiling was performed using an Ar⁺ etching. The S 2p XPS spectra (Fig. 4a) for the 5 wt% Zn(MEEA)2 electrolyte revealed peaks at 168.9 eV (ZnSO4 or ZHS, as previously reported)39, 167.3 eV (ZnSO3), and 161.8 eV (ZnS), which persisted after Ar⁺ etching for 210 and 420 s, corresponding to ~5 and ~10 nm depth, respectively. In comparison, only the peak at 168.9 eV is detected in the control, and in much lower quantities; the peak disappears after etching (Fig. 4b). The presence of ZnS as observed with the additive may result from the electrochemical reduction of SO42− as previously proposed in the literature19,20. This suggests that Zn(MEEA)2 promotes the in-situ formation of ZnS-based SEI, improving the Zn electrodeposition, as seen in the SEM images and cell performance. The presence of any organic SEI components originating from the Zn(MEEA)2 decomposition was studied by collecting C 1s XPS spectra. Aside from adventitious carbon (284.8 eV), peaks at 286.5 eV (C–O) and 289.6 eV (C = O) are observed. These peaks persist after Ar⁺ etching, suggesting contributions beyond sole hydrocarbon contamination. The persistence of these peaks when using 5 wt% Zn(MEEA)2 suggests an organic component to the SEI, potentially from anion decomposition. This hypothesis can be supported by the C 1s XPS spectra of 2 M ZnSO4, where peaks at 284.8 eV and 289.6 eV (C = O) disappear after Ar+ etching (Fig. S20). However, it must be noted that adventitious carbon complicates C 2s XPS analysis and we are wary of making any firm conclusions regarding the organic components of the SEI. Examining the Zn 2p spectra for 5 wt% Zn(MEEA)2 displays peaks for Zn2+ (2p1/2 = 1044.4 eV and 2p3/2 = 1021.9 eV), indicative of Zn-based species on the surface, i.e., ZnS and/or Zn-organic species (Fig. 4c). After Ar+ etching, this peak reduces in intensity to gradually reveal the metallic Zn peaks (2p1/2 = 1043.8 eV and 2p3/2 = 1021 eV), confirming the presence of an SEI on the Zn surface.Fig. 4X-ray photoelectron spectroscopy of Zn negative electrodes cycled 20 times at 0.624 mA.cm−2/2.5 mAh.cm−2.a S 2p XPS spectra for Zn negative electrodes (plated electrode) cycled in 2 M ZnSO4 (S 2p 3/2 and S 2p 1/2 in a 2:1 ratio with a splitting of ~ 1.18 eV) and b S 2p XPS spectra for Zn negative electrodes (plated electrode) cycled in 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with Ar+ etching of ~7 nm (S 2p 3/2 and S 2p 1/2 in a 2:1 ratio with a splitting of ~ 1.18 eV), c Zn 2p XPS spectra for Zn negative electrodes (plated electrode) in 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with Ar+ etching of ~7 nm, d Schematic of Zn metal surface in 2 M ZnSO4 and 2 M ZnSO4 + 5 wt% Zn(MEEA)2, e ZnS@Zn||ZnS@Zn stripping/plating for 2 M ZnSO4 cycled at 0.624 mA.cm−2/2.5 mAh.cm−2 f SEM and EDX of ZnS@Zn metal taken ex-situ after 1000 h. All battery cycling took place under ambient conditions. X-ray photoelectron spectroscopy of Zn negative electrodes cycled 20 times at 0.624 mA.cm−2/2.5 mAh.cm−2. a S 2p XPS spectra for Zn negative electrodes (plated electrode) cycled in 2 M ZnSO4 (S 2p 3/2 and S 2p 1/2 in a 2:1 ratio with a splitting of ~ 1.18 eV) and b S 2p XPS spectra for Zn negative electrodes (plated electrode) cycled in 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with Ar+ etching of ~7 nm (S 2p 3/2 and S 2p 1/2 in a 2:1 ratio with a splitting of ~ 1.18 eV), c Zn 2p XPS spectra for Zn negative electrodes (plated electrode) in 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with Ar+ etching of ~7 nm, d Schematic of Zn metal surface in 2 M ZnSO4 and 2 M ZnSO4 + 5 wt% Zn(MEEA)2, e ZnS@Zn||ZnS@Zn stripping/plating for 2 M ZnSO4 cycled at 0.624 mA.cm−2/2.5 mAh.cm−2 f SEM and EDX of ZnS@Zn metal taken ex-situ after 1000 h. All battery cycling took place under ambient conditions. To validate the stability of the SEI formed via Zn(MEEA)2, Zn||Cu and Zn||Zn cells were conditioned in 2 M ZnSO4 + Zn(MEEA)2 and then cycled in a base electrolyte of 2 M ZnSO4. In the Zn||Cu cell, both conditioning and reservoir steps were carried out with Zn(MEEA)2, after which the electrolyte was replaced with plain ZnSO4 (see “Methods” for details) (Fig. S21a). A Zn||Zn cell conditioned for 10 cycles in Zn(MEEA)2 electrolyte sustained over 600 h of plating/stripping after switching to 2 M ZnSO4. Furthermore, unlike the control Zn||Cu cell with 2 M ZnSO4, which failed before 49 cycles, this cell achieved stable plating/stripping with a CE of 96.5%. These results highlight the stability of the ZnS-based SEI formed by Zn(MEEA)2 (Fig. S21b). To ascertain the importance of a ZnS-based SEI on Zn electrodeposition, an artificial ZnS layer was formed onto a Zn surface with ex-situ stripping/plating in 2 M ZnSO4. The ZnS layer on Zn metal was achieved using an adapted method from literature40. Zn metal was added to a quartz tube with S8 powder and sealed under an inert atmosphere before being heated to high temperatures (> 300 °C). The presence of ZnS layer onto Zn was confirmed via EDX which showcased the presence of only Zn and S on the surface of the negative electrode (Fig. 4e). A ZnS@Zn||ZnS@Zn cell using 2 M ZnSO4 was cycled at 0.624 mA.cm−2/2.5 mAh.cm−2, such improved performance are similar to that of 2 M ZnSO4 + 5 wt% Zn(MEEA)2, confirming the role played by ZnS in the enhancement of Zn plating/striping performance (Fig. 4f). To shed light on the mechanistic formation of the ZnS SEI with the zinc oligoether carboxylates, the interplay between Zn electrodeposition and HER, both with and without Zn(MEEA)2, was studied using scanning electrochemical microscopy (SECM) in a generation collection mode (Fig. 5a)41. Briefly, Zn is platted onto a substrate during cyclic voltammetry scans, while a Pt microelectrode (Pt-ME) positioned at a ~ 10 µm above the substrate is held at a potential outside that of Zn electrodeposition (i.e., −0.5 V vs Ag/AgCl) to collect H2 generated at the Zn surface during the electrodeposition process (see details in “Methods”). During the first Zn electrodeposition, the HER current (itip) detected for 2 M ZnSO4 + 5 wt% Zn(MEEA)2 is doubled (20 pA) compared to that of the control 2 M ZnSO4 electrolyte (10 pA) (Fig. 5b & c). After three cycles, the HER current (itip) for 2 M ZnSO4 + 5 wt% Zn(MEEA)2 decreases to 5 pA. Meanwhile, no change to the HER current is detected for 2 M ZnSO4. This result indicates that the presence of Zn(MEEA)2 at the interface, as confirmed by the decreased EDLC measured upon addition of Zn(MEEA)2, results in the passivation of the Zn surface related to an excess of HER on the initial cycle. It would appear that the passivation is incomplete in SECM experiments; however, differences exist between the passivation across different cell configurations i.e., coin or Swagelock. Indeed, the SEI morphology and precipitation mechanism is highly dependent on stack pressure and the ratio between the active surface area and the volume of electrolyte.Fig. 5Scanning electrochemical microscopy measurements.a schematic for SCEM setup41, b SCEM data for 2 M ZnSO4 with isub bottom panel and itip top panel, c SCEM data for 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with isub bottom panel and itip top panel, d cyclic voltammetry of electrolytes at 5 mV/s on glassy carbon (WE), Pt wire (CE) and Ag/AgCl leakless electrode (REF) for 2 M ZnSO4 (green) 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple) and d schematic of EDL and subsequent reactions occurring on the Zn negative electrode surface. Scanning electrochemical microscopy measurements. a schematic for SCEM setup41, b SCEM data for 2 M ZnSO4 with isub bottom panel and itip top panel, c SCEM data for 2 M ZnSO4 + 5 wt% Zn(MEEA)2 with isub bottom panel and itip top panel, d cyclic voltammetry of electrolytes at 5 mV/s on glassy carbon (WE), Pt wire (CE) and Ag/AgCl leakless electrode (REF) for 2 M ZnSO4 (green) 2 M ZnSO4 + 5 wt% Zn(MEEA)2 (purple) and d schematic of EDL and subsequent reactions occurring on the Zn negative electrode surface. Considering that 2 M ZnSO4 is only mildly acidic (pH 3.8), the primary source of protons for the HER is H2O. For protons to be available, a sluggish water dissociation step (cleavage of an O-H bond, so-called Volmer step) is required to form H(ads) on the Zn surface (Eq. (1)) prior to forming H2(g). This reaction has been widely shown to have kinetic limitations where the direct reduction of protons, either through lower pH or proton donors, provides a more favourable HER42–44. More importantly, the dissociation step generates OH− (Eq. (1)), promoting the precipitation of ZHS, as previously discussed39.1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${H}_{2}O+{e}^{-}\to \,{H}_{{abs}}+\,{{OH}}^{-}$$\end{document}H2O+e−→Habs+OH− Intriguingly, the pKa of the COO− group from Zn(MEEA)2 is determined to be 4.5 by titration, indicating that MEEA anion exists predominantly (83.3%) in its protonated form in 2 M ZnSO4, as determined by the Henderson–Hasselbalch equation (Fig. S22 and Eq. (2)). The slightly elevated pKa compared to acetic acid (pKa ~4.76) can be explained by the electron-donating ether chain, which acts to stabilise the conjugate base.2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${pH}={pKa}+\log \frac{\left[{{MEEA}}^{-}\right]}{\left[{HMEEA}\right]}$$\end{document}pH=pKa+logMEEA−HMEEAHence, HMEEA readily supplies protons to the Zn surface (Eq. (3)), supporting the increase in HER current observed in the SECM data during the first cycle.3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$({MEEA}){H}^{+}+{e}^{-}\to \,{H}_{{abs}}$$\end{document}(MEEA)H++e−→Habs This result is in line with EDLC estimation that suggests that MEEA is preferentially located at the interface, displacing water from the surface. The viability of this conclusion is well-supported by fundamental HER studies, which demonstrate that -COOH groups can act as proton donors and buffers for the HER, preferentially promoting H⁺ reduction over H2O reduction42,43. The favourable reduction of H⁺ promotes the formation of adsorbed H species and/or H2, which subsequently accelerates the reduction of SO₄²⁻ to S²⁻. The latter then combines with Zn2+ to form ZnS in the SEI. (Eq. (4)).4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{{{\rm{SO}}}}}_{4}}^{2-}+{{8}{{{{\rm{H}}}}+}}_{{{{\rm{abs}}}}}\to {{{{\rm{S}}}}}^{2-}+{{{{\rm{H}}}}}_{2}{{{\rm{O}}}}\to {{{{\rm{Zn}}}}}^{2+}+{{{{\rm{S}}}}}^{2-}\to {{{\rm{ZnS}}}}$$\end{document}SO42−+8H+abs→S2−+H2O→Zn2++S2−→ZnS This observation would seemingly match prior descriptors for the formation of a ZnS-based SEI in the presence of large concentration of protons. Our work nonetheless deviates from previous strategies relying on the use of superacidic conditions and demonstrate that ZnS-based SEI can be achieved via a proton-donating group under mild pH conditions. Furthermore, our SECM study confirms that ZnS stabilises the Zn metal surface by suppressing further HER, thereby ensuring long-term negative electrode stability (Fig. 5d). Overall, these findings with the SCEM in tandem with the EDLC highlight the synergistic effect of the additive. The carboxylic acid group provides localised acidity, which enables the formation of the SEI, while the long ether chains create a hydrophobic domain on the Zn surface to mitigate the HER. To ascertain a potential application for the zinc oligoether carboxylates, Zn(MEEA)2 was used as an additive in a full cell. Our previous work focused on the development of rechargeable Zn||MnO2 batteries. Our study highlighted that urea is essential for stabilising MnO2 by suppressing Mn electro-dissolution, thus preventing capacity decay3,45,46. However, urea itself is unable to stabilise the Zn negative electrode, where dendrites still persist and thus performance in Zn||Zn with 2 M ZnSO4 + 10 M urea matches the ~200 h stripping/plating obtained with 2 M ZnSO4 (Fig. 6a)47. However, upon addition of 5 wt% Zn(MEEA)2 to 2 M ZnSO4 + 10 M urea, stripping/plating is extended to 1000 h (Fig. 6a). This increase in performance is attributed to the ZnS-based SEI, confirmed via the XPS S 2p spectra (Fig. S23). This indicates that urea does not impact the aforementioned mechanism for the formation of ZnS. Similarly to the Zn||Zn testing, Zn||MnO2 full cells were cycled in Swagelok-type cells using low current densities with a high mass loading of the positive electrode composite (6.25 mg.cm−2 of MnO2). This approach offers a more practical evaluation of the system, as high current densities (> 2 C) or low mass loadings ( < 2 mg mg.cm−2 of MnO2) often obscure the typical issues associated with MnO2 electrodes. When cycling with 2 M ZnSO4 + 10 M urea dendrites occur, leading to a short circuit after only 30 cycles at 0.624 mA.cm−2/2.5 mAh.cm−2 (0.25 C) (Fig. 6b, c). Meanwhile, adding 5 wt% of Zn(MEEA)2, extends the cycle-life almost tenfold to 200 cycles. These results in Zn||Zn and Zn||MnO2, highlight the synergy between the additives, where Zn(MEEA)2 stabilises the Zn metal negative electrode, while urea stabilises the MnO2 positive electrode, resulting in enhanced full cell performance under practical conditions. This beneficial effect is revealed by observing that the cell impedance is greater with the additive, which is the result of an increased ohmic drop from the Zn negative electrode, as revealed by three-electrode electrochemical impedance spectroscopy and galvanostatic testing (Fig. S24). Finally, these promising results were corroborated by measurements in pouch cell configuration (Fig. S25a), utilising larger loadings of active materials, approximately 25× increase (10.5 mg.cm−2 i.e., total loading = ~158 mg in pouch cell vs. 6.25 mg.cm−2 i.e., total loading = ~6.25 mg in Swagelock configuration). The pouch cell using 2 M ZnSO4 + 10 M urea electrolyte fails after less than 20 cycles. However, relatively steady cycling, albeit with a constant capacity fading of less than 0.2% per cycle, was obtained for more than 90 cycles when adding 5% Zn(MEEA)2. This performance enhancement can be attributed to persistent HER in the control electrolyte, observed via the swelling of the pouch cell upon cycling, meanwhile, no such swelling was observed upon addition of 5% Zn(MEEA)2 (Fig. S25b).Fig. 6Use of Zn(MEEA)2 with Urea-based electrolyte & Zn||MnO2 cycling in coin and pouch cell.a Zn||Zn cell cycled with 2 M ZnSO4 + 10 M Urea (Grey) and 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 0.624 mA.cm−2/2.5 mAh.cm−2. b Zn||MnO2 Swagelok cycling for 2 M ZnSO4 (grey) & 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 0.624 mA.cm−2/2.5 mAh.cm−2 (points taken every 5 cycles) c corresponding cycling curves. d Zn||MnO2 Pouch cell cycling for 2 M ZnSO4 (grey) & 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 1.25 mA.cm−2/2.5 mAh.cm−2. All cycling took place under ambient conditions. Use of Zn(MEEA)2 with Urea-based electrolyte & Zn||MnO2 cycling in coin and pouch cell. a Zn||Zn cell cycled with 2 M ZnSO4 + 10 M Urea (Grey) and 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 0.624 mA.cm−2/2.5 mAh.cm−2. b Zn||MnO2 Swagelok cycling for 2 M ZnSO4 (grey) & 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 0.624 mA.cm−2/2.5 mAh.cm−2 (points taken every 5 cycles) c corresponding cycling curves. d Zn||MnO2 Pouch cell cycling for 2 M ZnSO4 (grey) & 2 M ZnSO4 + 10 M Urea + 5 wt% Zn(MEEA)2 (Green) at 1.25 mA.cm−2/2.5 mAh.cm−2. All cycling took place under ambient conditions. In this study, rationale design of a family of salt additives is introduced for improved Zn-based batteries, namely zinc oligoether carboxylate salts Zn(MA)2, Zn(MEA)2, and Zn(MEEA)2. Adding 5 wt% of these salts is found to significantly enhance Zn electrodeposition, leading to improved cycle life and coulombic efficiency in both Zn||Zn and Zn||Cu cells, with the best-performing Zn(MEEA)2 achieving over 1000 h of plating/stripping, with a CE% of 99.9%. The performance improvement is attributed to the formation of a ZnS-based SEI, confirmed via XPS, SEM and XRD. By using SECM with titration techniques, we demonstrate that protonated oligoether carboxylate chains create a localised acidic environment on the Zn surface, favourable for the reduction of SO42− into S2−, precipitating ZnS that effectively passivates the electrode. When applied to a previously developed urea-based electrolyte for Zn||MnO2 full cells, Zn(MEEA)2 compensates for urea’s inability to mitigate dendrites on the Zn negative electrode while preserving its beneficial effects on the MnO2 positive electrode. These findings facilitated the fabrication of a pouch cell, representing a key step toward technological scale-up. Bridging the gap between lab-scale promise and real-world applicability remains a significant challenge. Scaling up the synthesis of these salts and eliminating the need for an inert atmosphere will require targeted synthetic expertise and further development. While the cell performance in this study is strong, and the testing conditions offer a more realistic evaluation than typically seen in similar research, optimising a practical cell remains an ongoing challenge. Going forward, setting a standard when it comes to cycling conditions would enable fair comparisons between different strategies. There is a delicate balance required in optimising the fabrication of a practical cell and electrolyte formulation suited to this format, ideally capable of cycling across a wide range of temperatures and conditions. While the cell format of a Zn||MnO2 battery remains unknown, the ideal electrolyte formulation does too. Nevertheless, we believe this in-depth assessment of the chemistry of the Zn-electrolyte interface has enabled a clearer comprehension of the nucleation-growth of Zn dendrites, while positioning zinc oligoether carboxylate salts as a promising advancement in AZMB technology. We hope that these salts can help to drive the development of green and sustainable next-generation battery systems based on the Zn metal negative electrode. Thermogravimetric analysis/difference scanning calorimeter The thermal stability was analysed by TGA/DSC using a Mettler Toledo TGA/DSC3 + LF/1100. Around ~10 mg of the sample was put in an aluminium crucible and heated at a rate of 5 °C/min from ambient temperature to 400 °C under air. Electrolyte characterisations Electrolyte preparation for zinc oligoethers For electrolytes using the oligoethers as a primary salt, 1 M of Zn(MA)2, Zn(MEA)2 and Zn(MEEA)2 was dissolved into Milli-Q ultrapure water. For electrolytes using the oligoethers as an additive, 2 M ZnSO4 was prepared by dissolving ZnSO4 (Alfa Aesar, 99%) salt into Milli-Q ultrapure water. The mixtures were stirred for 5 min. The desired amount of the additive of 2.5 wt%, 5 wt% or 10 wt%, was added to a 2 M ZnSO4 blank electrolyte and the solution was vortexed for 1 min. For the electrolytes containing urea, 10 M of urea (99.9% Sigma Aldrich) was dissolved in Milli-Q ultrapure water, to which 2 M ZnSO4 was added. Zn(MEEA)2 was then added at 5 wt% to the solution. Ionic conductivity measurements The ion conductivities (σ) were measured using a Mettler-Toledo SevenCompact S230 conductivity meter with a 12 mm InLab® 710 Cond probe with 4 Pt poles conductivity cell ( ± 0.5%). IR spectroscopy FTIR spectra were acquired in ATR mode on a NicoletTM iSTM 5 FTIR spectrometer (Thermo Fisher Scientific) in the range of ~500–4000 cm−1. Electrochemical measurements Zn||Zn symmetric cells using a PFA Hardware Swagelok cell were assembled with two zinc metal electrodes (Zn disk 1 cm Ø, 0.25 mm thickness), which were untreated before use. Two Whatman glass fiber separators (GF/D, 1.27 cm Ø, 2.7 μm pore size, 0.67 mm thickness) were added between the two electrodes along with 100 µL of electrolyte, carefully incorporated using a micropipette. The Swagelok cells were sealed with parafilm© to ensure a tight seal. The cycling rate of the cells were 0.624 and 2.5 mA.cm−2 with a capacity of 2.5 mAh.cm−2, which corelates to 0.25 and 1 C. CA was carried out using Zn||Zn cells at −150 mV. For the cells that were first conditioned using 2 M ZnSO4 + 5 wt% Zn(MEEA)2 based electrolyte, a 10 mm ring which was place between the two electrodes with 300 µL electrolyte solution during the precycling; this facilitated the removal of the electrodes and suppressed the need for using a separator before running with the control electrolyte with separators. This large amount of electrolyte and the separation between both electrodes explains why the overpotential is slightly higher during the conditioning. Cycling of the cells were not carried out in climatic/environmental-controlled chambers. Zn||Cu coin cells were prepared using copper metal foil electrodes (99.99%, Cu disk 1 cm Ø, 0.1 mm thickness) and zinc metal electrodes (Zn disk 1 cm Ø, 0.25 mm thickness) which both were untreated before use. Two Whatman glass fiber separators (GF/D, 1.27 cm Ø, 2.7 μm pore size, 0.67 mm thickness) were added between the two electrodes along with 100 µL of electrolyte. The cycling conditions were adapted from Xu et al.30. Briefly, asymmetric Zn | |Cu cells were cycled with an initial conditioning cycle (4 mAh.cm−2) used to plate and strip Zn from the Cu working electrode, using a cut-off voltage of 0.5 V at a current density of 1 mA.cm−2 after which a reservoir of 4 mAh.cm−2 was plated at a current density of 1 mA cm−2 on the Cu to provide a controlled and limited source of Zn (Qr). Subsequently, cells were cycled 49 times using a fixed capacity of 1 mAh.cm−2 (Qc) at a current density of 1 mA cm−2, followed by stripping to a cut-off voltage of 0.5 V, ensuring that all removable Zn, including the initial reservoir was stripped (Qs). The CE% is then calculated using:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${CE}=\,\frac{49{Q}_{c}+{Q}_{s}}{49{Q}_{c}+{Q}_{r}}$$\end{document}CE=49Qc+Qs49Qc+Qr Cycling of the cells were not carried out in climatic/environmental-controlled chambers. Scanning electrochemical microscopy SECM experiments were performed using a CHI 910b bi-potentiostat (CH Instruments). A platinum ultramicroelectrode (Pt UME, 25 µm diameter) was used as the tip working electrode, while a zinc electrode served as the substrate working electrode. A platinum wire was employed as the counter electrode, and a leakless Ag/AgCl electrode was used as the reference electrode. The Pt UME was positioned approximately 10 µm above the substrate using negative feedback approach curves with O₂ as an irreversible redox mediator (ref). The substrate generation/tip collection (SG/TC) mode was employed to detect H₂ generated at the zinc electrode. A potential of –0.5 V vs Ag/AgCl was applied to the Pt UME, providing sufficient overpotential for efficient H₂ oxidation. Capacitance measurements Initial zinc stripping/plating measurements was performed in an electrochemical glass cell using a PTFE embedded glassy carbon disc (3 mm diameter, Pine Research Instrumentation) as working electrode, Pt wire as counter electrode and Ag/AgCl leakless electrode was used as a reference. ~5 mL of electrolyte was used for each measurement. Electric double-layer capacitance (EDLC) was measured using cyclic voltammetry (CV) of Zn||Zn symmetric cells in various electrolytes at sweep rates of 4–14 mV/s within the non-Faradaic region ( ± 15 mV). EDLC was calculated using the equation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{\rm{C}}}}={{{{\rm{i}}}}}_{{{{\rm{c}}}}}/{{{\rm{v}}}},$$\end{document}C=ic/v,where C is the capacitance, ic is the double-layer current, and v is the scan rate. The capacitance was determined from the slope of ic versus v, with ic defined as half the current difference at 0 V during the forward and reverse scans:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{{\rm{i}}}}}_{{{{\rm{c}}}}}=({{{{\rm{i}}}}}_{{0{{{\rm{V}}}}}_{+}}-{{{{\rm{i}}}}}_{{0{{{\rm{V}}}}}{-}})/2$$\end{document}ic=(i0V+−i0V−)/2Zn||MnO2 full cells were prepared using a PFA Hardware Swagelok cell. CMD-γ-MnO2 was synthesised according to ref. 45. CMD-γ-MnO2 was synthetised by adding 50 ml of 0.6 M of Na2S2O8 (Sigma Aldrich, 99%) into a round bottom flask containing 125 ml of 0.2 M MnSO4 (Sigma Aldrich, 99.9%) under heating and magnetic stirring. The precipitate was washed with MilliQ water, filtered through a cellulose ester filter and dried at 100 °C under vacuum for 24 h. MnO2 was mixed and grinded with super-p carbon in a ratio of 1:1. Ohmic drop measurements were performed by subtracting the first recorded potential of the subsequent step from the final potential of the preceding charge or discharge step, and dividing the result by the change in current. The measurement was taken over a time interval of 0.6 ms. 12.5 mg (6.25 mg of CMD-γ-MnO2 and 6.25 mg of Ketjenblack EC-600JD (KB) carbon (Cheemtree)) of powder was used as the positive electrode, meanwhile zinc metal acted as the negative electrode (Zn disk 1 cm Ø, 0.25 mm thickness). Two Whatman glass fibre separators (GF/D, 1.27 cm Ø, 2.7 μm pore size, 0.67 mm thickness) were added and soaked with 100 µL of electrolyte. The Swagelok cells were sealed with parafilm© to ensure a tight seal. The cycling rate of the cells were 0.624 and 1.25 mA.cm−2 with a capacity of 2.5 mAh.cm−2. Pouch cells (3 × 5 cm) were prepared by mixing MnO2 powder (Thermo Scientific), Ketjenblack EC-600JD (KB) carbon (Cheemtree), and Super Conductive Carbon Black P at a weight ratio of 70:15:15 for the single-sided positive electrode. Then, 0.5 wt% of polytetrafluoroethylene (PTFE) binder dispersion (Sigma-Aldrich, 60 wt% in H2O) was added to the mixture. The resulting paste was rolled into a thin film and cut into electrodes of 3 × 5 cm, with a mass loading of 15 mg.cm−2 i.e., active material loading = 10.5 mg.cm−2. The electrodes were then pressed onto stainless steel mesh (Alfa Aesar, 80 mesh woven from 0.14 mm dia wire, type 316) using a hand press. A single glass fiber membrane (GF/D, 1.27 cm Ø, 2.7 μm pore size, 0.67 mm thickness) was used as the separator, soaked with 3–4 mL of electrolyte solution (2 M ZnSO4/10 M urea/5 wt% Zn(MEEA)2). The negative electrode consisted of Zn foil (70 µm thickness, 99.995% purity). The pouch cells are single layer. Pouch cells were assembled under ambient air atmosphere for electrochemical measurements. The voltage window was set from 0.55 V to 1.8 V, with a specific current corresponding to 98 mA.g−1 of MnO2, using a modular battery cycler (Biologic BCS 810). Cycling of the cells were not carried out in climatic/environmental-controlled chambers. The applied pressure for the pouch cells during cycling was 1 MPa. Scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) Images were taken with a JEOL 7800 F Prime at an acceleration voltage of 6.0 kV and 10 mm working distance and energy dispersive X-ray spectra in STEM-mode (STEM-EDX) were collected with a probe-corrected Titan Themis Z electron microscope operated at 200 kV and equipped with a Super-X EDX detector. X-ray diffraction (XRD) patterns were acquired on a Bruker D8 Advance diffractometer with Bragg–Brentano geometry, a Cu-Kα source (λ1 = 1.54056 Å, λ2 = 1.54439 Å) and a LynxEye detector. X-ray photoelectron spectroscopy (XPS) measurements were conducted on Thermo Fisher Scientific K-alpha instrument with an Al X-ray source. Pristine and ex situ powder samples were pressed into a copper holder and mounted in a vacuum transfer chamber in an argon-filled glovebox to avoid exposure to air and moisture. Data fitting and analysis was performed using Avantage software. Methods Synthesis Zn(MA)2 was prepared by dissolving 2-(2-methoxyethoxy)acetic acid (TCI, 99%) in water to which Zn5(CO3)2(OH)6 (Sigma Aldrich, 99%) was added until the pH of the solution reached 7. The resulting solution was distilled under vacuum to obtain the zinc (2-(2-methoxyethoxy)acetate)2. Zn(MEA)2 was prepared by dissolving 2-[2-(2-methoxyethoxy)ethoxy]acetic acid (TCI, 99%) in water to which Zn5(CO3)2(OH)6 (Sigma Aldrich, 99%) was added until the pH of the solution reached 7. The resulting solution was distilled under vacuum to obtain the zinc (2-[2-(2-methoxyethoxy)ethoxy]acetate)2. Zn(MEEA)2 HMEEA is prepared in a Ar-filled glovebox by dissolving sodium metal (0.132 mol) in excess TEGME (TCI, 99%) (2-(2-(2-Methoxyethoxy)ethoxy)ethanol) (50 mL), after which chloroacetic acid (Sigma Aldrich, 99.99%) (0.067 mol) dissolved in TEGME (10 mL) is added dropwise into the solution and is reacted for 16 h. Excess TEGME is removed via distillation after which the solution is treated with an aqueous solution of orthophosphoric acid (0.098 mol) (Sigma Aldrich, 99.99%). Dichloromethane (DMC) (VWR, 99%) was added to the solution, resulting in two phases. The aqueous phase was washed further with DMC and the organic phases were combined and dried using magnesium sulphate. This solution is then distilled under vacuum to yield the HMEEA22. The Zn(MEEA)2 was prepared via cation exchange reaction in water using Zinc carbonate basic (Zn5(CO3)2(OH)6) (Sigma Aldrich, 99%). 2-[2-(2-methoxyethoxy)ethoxy]acetic acid was dissolved in water to which Zn5(CO3)2(OH)6 was added until the pH of the solution reached 7. The resulting solution was distilled under vacuum to yield the zinc (2-[2-[2-(2-methoxyethoxy)ethoxy]ethoxy]acetate)2. Titration of HMEEA 100 mM of HMEEA was used. The solution was brought down to pH 2 using HCl (37%, Carlo Erba) 200 mM solution, then 1 mL aliquots of NaOH ( > 98% pellets, Sigma Aldrich) 200 mM solution was added gradually. NMR measurements To confirm a successful synthesis of the salts 1H and 13C liquid-state NMR spectra was carried out recorded in a Bruker 7.046 T (300 MHz for 1H) Advance III NMR spectrometer mounted with a 5 mm HX(F) probe head, using CDCl3 (Sigma Aldrich, 99.99%) as the solvent. Electrospray Ionisation mass spectrometry (ESI-MS) MS analysis was carried out by infusing the sample in the electrospray source of a Synapt G2-S (Waters, SN: UEB205) mass spectrometer operating in positive or negative mode. Capillary and cone voltage were respectively 3000 V and 30 V. Source and desolvation temperatures were respectively 250 °C and 100 °C. Mass spectra were acquired over a range of 50–1500 Da. Elemental analysis 15 mg of the sample was placed in a tin capsule and was transferred into an inert gas chamber (zero blank) under Helium flow. After which the catalytic combustion of the capsule and sample at a temperature of 1150 °C was carried out. Followed by a reduction of the combustion gases over hot copper in a second furnace at 850 °C. The gases formed for analysis N2, CO2, H2O, and SO2 remain in a Helium carrier gas stream. Separation of the gas mixture in a TPD column (Elementar patent), “temperature-programmed desorption column”. Signal detection by a TCD katharometer “thermal conductivity detector” for C, H, and N nuclei. Supplementary information Supplementary Information Peer Review File Supplementary Information Peer Review File Source data Source Data Source Data