Abstract
Eutectic electrolytes hold promise for aqueous zinc metal batteries in sustainable energy storage chemistries, yet improvement from perspective of molecule configurational engineering are ambiguous. Herein, we propose design strategy of increasing asymmetric molecular geometry in organic ligands to regulate frustrated coordination and disordered structure for eutectic electrolytes toward enhanced zinc metal batteries. The introduced asymmetry in eutectic component gives rise to relatively weak coordination strength and configurational disorder interaction among cation‐anion‐ligand, leading to suppressed local aggregation, steady eutectic phase and improved Zn2+ diffusion kinetics. Such highly frustrated coordination state also enables disruption of hydrogen bonding network and reinforcement of anion participation, which results in confined side reactions, decreased water activity and the formation of inorganic‐enriched solid electrolyte interphase. In comparison to highly symmetric ligands, asymmetric ligand‐involved eutectic electrolytes with configurational disorder deliver high Coulombic efficiency of 99.4 %, stabilized Zn plating/stripping of 5000 h and impressive rate capability even under harsh conditions such as small N/P, low temperature. The rationale in this work advances the deep understanding of asymmetric molecular engineering in eutectic electrolytes and showcases suitability of frustrated coordination to achieve high‐performance zinc metal batteries.
Keywords: Frustrated coordination, Ligand asymmetry, Eutectic electrolyte, Zinc batteries
Large coordination frustration driven by molecular asymmetry of ligands and disordered cation‐ligand‐anion‐water eutectic coordination can lead to relatively enhanced chemical stability, weak coordination strength and promoted mass mobility in eutectic electrolytes for high‐performance zinc batteries.

Introduction
Rechargeable zinc metal batteries (ZMBs) have been revitalized as a viable technology for efficient energy storage in electrification of vehicles or smart grids. [1] Despite its intrinsic safety and low cost, there is still a giant leap towards widespread commercialization of zinc‐based batteries, which mainly hinges on the formation of side reactions and zinc dendrite, originally stemming from the free water in aqueous electrolytes.[ 2 , 3 , 4 ] In this regard, new research direction within the various solvation structure regulation and chemical components selection in electrolyte engineering is expected to improve reversibility and durability of zinc batteries.[ 5 , 6 , 7 ]
This motivates the investigation on how the various types of composition or the number of water substituents in electrolyte impact the functional properties including charge transfer, ion conduction and solid electrolyte interface (SEI), et.al.[ 8 , 9 , 10 ] Modification strategies such as introduction of co‐solvent/additive, change of anion/salt concentration, construction of quasi‐hydrogel have been explored to regulate the dynamics of electrode interfacial deposition and influence the local interactions within solvation structure.[ 11 , 12 ] An important aspect here is to reasonably utilize eutectic electrolytes associated with the presence of multiple principal components via intermolecular interactions (e.g. Lewis acid‐base interaction, H‐bonding interaction and van der Waals interaction).[ 13 , 14 ] Given the complementary merits of feasible preparation and high thermal/electrochemical stability, eutectic electrolytes are proven to be conducive to sustainable Zn plating/stripping. [15] Despite these advances, current state of knowledge of eutectic chemistry is still far from complete. Various eutectic configurations with appropriate compatibility of zinc salt and primary ligand have been developed, molecular geometry of the organic ligands is rarely taken into account to alter cation‐anion‐ligand cluster coordination.[ 16 , 17 ] One essential design methodology is to determine whether the changes of molecular asymmetry configuration in eutectic ligands can influence the electrolyte behavior or battery performance.[ 18 , 19 ]
Differing from mixing solvents which is a common method to compensate deficiencies of the individual component,[ 20 , 21 , 22 ] by coordinating local organic ligands with molecular asymmetry and configurational disorder in eutectic Zn2+ regime, a more complex and dynamic interaction structure is expected to form. It is acknowledged that the complex electrolyte structure with disordered nature or increased entropy can benefit the kinetic and thermodynamic properties such as conductivity, diffusivity and redox stability in zinc batteries.[ 23 , 24 , 25 ] Eutectic liquids often leave a kinetic gap of molecule/ion dissociation and charge motion due to high eutectic affinity in ordered structural eutectic constitutions.[ 26 , 27 ] For asymmetric ligand‐participated systems with frustrated coordination and higher entropic driving force, the spatial transfer effect of intermolecular interactions in eutectic electrolyte will be significantly hindered and result in relatively weak coordination strength, leading to the reduced ordered arrangement of eutectic structure and enhanced mass transfer. [28] While the regulation of molecular asymmetric distortion in local organic ligands with higher configurational frustration as a design knob has largely gone unnoticed for eutectic electrolytes. [29] Thus, we believe introducing asymmetric ligands with coordination disorder may contribute to high frustration in eutectic electrolytes and associated entropy driving thermodynamic equilibrium to favor suitable zinc‐ligand interactions, hydrogen bonding network and anion‐involved binding for zinc batteries.
In this study, we report a high frustration eutectic electrolyte concept by regulating local asymmetry of molecular ligands, which essentially leads to configurational disorder in eutectic coordination and enhanced reversibility of zinc electrode. Through a host of systematical characterization and simulation, we find that introducing highly asymmetric molecular geometry for eutectic ligands can result in a richer variation and distortion of frustrated coordination structure, which is vital in determining the significant influence on ion transport mobility, cation‐ligand‐anion interaction and hydrogen bonding network of eutectic electrolytes. Specifically, the mass transfer and ion motion kinetics are promoted as easy aggregation and integrating of component distribution is hindered in eutectic mixture with configurational disorder. Meanwhile, the inductive effect on asymmetric ligand endows reduced negative charge transfer and lower binding energy with Zn2+, thereby enhancing anion association in coordinated electrolyte structure. Additionally, the breakdown of original hydrogen bonding network in eutectic regime with high structural disorder simultaneously suppresses hydrolysis‐involved side reactions and increases diffusion dynamics in harsh condition. As a proof of concept, three eutectic ligands including asymmetric N‐methylacetamide (MA), partially symmetric acetamide (AA) and highly symmetric urea (UA) are subsequently investigated. Results show that asymmetric ligand MA‐involved eutectic electrolytes produce relatively loose cation‐ligand‐anion eutectic constitutions with maximized degree of frustration and weak coordination strength, while still preserving the characteristic eutectic properties compared to AA and UA. The consequence is the weakened interionic order pairing, accelerated Zn2+ diffusion kinetics and formation of inorganic‐rich solid electrolyte interphase layer, which mitigates dendrite formation during high‐rate cycling and allows for uniform, dense deposition morphology. Collectively, the batteries with highly frustrated eutectic electrolytes can deliver stable cycling for 5000 h and high Coulombic efficiency (CE) of 99.4 %. Stable operation and excellent rate capability are exhibited in Zn//VO2 full cells with small N/P ratio (5.4), low temperature (−20 °C) and assembled pouch cell. The work provides insights into selecting optimal ligand configuration and designing disorder coordination‐driven eutectic electrolytes toward advanced ZMBs.
Results and Discussion
Eutectic Behavior upon Asymmetric Ligand and High Frustration
One route to increase frustrated coordination and configurational disorder is increasing the eutectic ligand asymmetry, analogous to enriching the diversity of components in high‐entropy materials system. Structural asymmetry is one of the most important geometry characteristics of molecules, and it will affect or determine many properties such as spatial arrangement, melting point, intra/intermolecular force. For the electrolyte system with asymmetric ligands, the affected performance can be described from two aspects: kinetics and thermodynamics. In general, the intermolecular interactions between asymmetric molecules are greatly reduced by unfavorable arrangement or distance. Thermodynamically, the enthalpy change during phase transition is decreased along with more difficult orientation or arrangement of asymmetric molecules. In terms of kinetics, it is not conducive to the process of phase transition between order and disorder. According to Carnelley's rule, the greater entropy change can be produced in the phase transition process of asymmetric systems. [30] In this regard, we systematically regulate the molecular asymmetry of ligand in eutectic electrolytes for high‐performance ZMBs. We conjecture that large coordination frustration driven by disordered cation‐ligand‐anion‐water eutectic constitutions will lead to relatively enhanced chemical stability, weak coordination strength and promoted mass mobility/ion transport for eutectic electrolytes (Figure 1a).
Figure 1.
a) Schematic diagram of the frustrated coordination strategy. b) Surface electrostatic potential maps for Zn(ClO4)2‐UA, Zn(ClO4)2‐AA and Zn(ClO4)2‐MA. c) Computation of charge transfer between zinc and UA, AA and MA. d) FTIR spectra of Aqua, LAE, MAE and HAE. e) Digital images of observing the eutectic melting process. f) DSC curves of Aqua, LAE MAE and HAE. g) Ionic conductivity and viscosity properties of Aqua, LAE, MAE and HAE. h) SAXS profile shows decreased clustering with increased molecular asymmetry. i) Temperature coefficients, signified by the slopes of the lines, show that HAE has the highest structure entropy.
The molecular asymmetry level of selected eutectic ligands is first summarized to determine the orientation arrangement dynamics. In the common eutectic components, N‐methylacetamide (MA) with methyl group attached to amide group is described as C1 point group in high molecular asymmetry and structure frustration. While acetamide (AA) with Cs point group indicates partially asymmetric structure and urea (UA) is similar to H2O with C2V point group, constructing by highly symmetric two amino groups and ordered molecular orientation. The cation‐ligand interaction was then computed via density functional theory (DFT) to predict disorder of possible eutectic systems with Zn(ClO4)2 salt. As revealed in Figure 1b, the electrostatic distribution evidenced an obviously asymmetric electrostatic potential (ESP) mapping in the coordinated MA when compared to UA and AA. The charge transfer from amide group was 0.815, 0.812, and 0.773 e for UA, AA and MA, respectively, which matched the binding energy results, confirming the stronger deviating and enhanced frustration in MA‐involved eutectics when incorporating Zn2+ Ions (Figure 1c, Figure S1).
It is reckoned that eutectic constitutions can generally be altered by modulating the asymmetry or molecular orientation, which is accompanied by variations in the chemical and physical characteristics of eutectic systems. Starting with the baseline electrolyte of 1 M Zn(ClO4)2⋅6H2O aqueous electrolyte (Aqua), we first prepared hydrated deep eutectic electrolytes with various ligands MA, AA and UA with different molecular asymmetry selection criterion and denoted the corresponding eutectic electrolytes as HAE, MAE and LAE according to the asymmetric degree from high to medium to low (Figure S2). Meanwhile, the formulation of two contrastive eutectic HAEs with lower and higher MA contents was also examined and named HAE‐1 and HAE‐6. HAE can remain steady for six months without any phase separation. However, the long‐term eutectic stability cannot keep in LAE as precipitation is detected after one week of resting (Figure S3). FTIR spectra further provide insights into eutectic coordination structure. The broad O−H stretching vibration from H2O at 3000–3500 cm−1 in Aqua becomes one narrowed peak in HAE and two separated peaks in LAE, MAE, corresponding to NH group for MA and NH2 for UA, AA (Figure 1d). The characteristic peaks at 1421, 1380 cm−1 attributed to CH3 gradually arise upon adding higher ratio of MA and following order of HAE‐1, HAE and HAE‐6, which demonstrates gradual ligand participation in coordination cluster (Figure S4).
Accordingly, the intermolecular interactions are adjusted in HAE, MAE and LAE, which is reflected as a change in eutectic behavior toward melting phase. As displayed in Figure 1e, the phase transformation from solid to liquid states under combination of eutectic mixture with selected asymmetric ligands are monitored in digital images. Behaving like bulky crystal, LAE still exhibits a non‐uniform melting phase at 55 °C and the majority of component remains in solid state, which in turn requires more formation energy to produce eutectic character. For MAE, a large fraction of mixture is engaged in the eutectic network and exists in liquid zone. Remarkably, a much lower temperature is recognized at 45 °C for HAE, with eutectic framework closer to complete liquid phase, guaranteeing stable eutectic coordinating structure. As further illustrated in differential scanning calorimetry (DSC) measurement, compared to the observable exothermic peak in Aqua and LAE, only single endothermic platform appears with lowest glass transition temperature (Tg) of −105.5 °C in HAE, confirming the eutectic behavior and more stable intermolecular forces between moieties (Figure 1f, Figure S5). In addition, the required heat values of LAE, MAE and HAE during the phase transition are 0.980, 0.872 and 0.781 J/K⋅g, respectively, which further implies the weakest intermolecular forces and coordination frustration for HAE. Walden plots are then probed to study the mass transport kinetics (Figure S6). With the uplifted frustration and asymmetry intensity in eutectic ligand, the ion mobility of HAE tend to rise up as sub‐ionic conduction manner due to the disrupted internal order coupling of cation‐ligand‐anion cluster. [31] Consequently, higher ionic conductivity and moderate level of viscosity are achieved in HAE with promoted wettability evidenced by slightly decreased contact angle when comparing to LAE and MAE, which validates the accelerated Zn2+ diffusion motion via largely increased frustration and asymmetry (Figure 1g). It is known that high concentration and cation/solvent ratio may induce salt aggregates and contact ion pairs for HAE‐1. Alternatively, large content of organic eutectic MA may also substantially increase viscosity and reduce ionic activity in HAE‐6. In this regard, the proposed eutectic model in HAE suitably balances the component correlations between salt and ligand, contributing to the enhanced ionic mobility (Figure S7).
The subtle microscopic disordered structure in HAE was then investigated using small‐angle x‐ray scattering (SAXS) (Figure S8). Figure S9 illustrates a typical power‐law behavior at low scattering vector Q, which is correlated with homogeneous fractal structure in simulated scattering profile. There is a higher population of smaller clusters in HAE compared with LAE and MAE, which is related to smaller hydrodynamic coordination radius (Figure 1h). For Guinier plot, HAE is characterized as higher degree of deviation from linear region compared to LAE and MAE, representing the possible larger number of disordered configurations (Figure S10). Additionally, the temperature coefficient related to entropy change of electrochemical reaction is estimated in non‐isothermal cells and guided by the equation inserted in Figure 1i: where E is the equilibrium cell voltage, T is the temperature, n is the number of electrons transferred in the reaction, F is Faraday's constant. [32] It is observed that temperature coefficient varies larger in HAE than MAE and LAE, manifesting higher electrolyte frustration and structure disorder induced by asymmetric ligand coordination.
Investigation of the Hydration Structure
Let us first address the term (i). Influence of frustration‐driven eutectic coordination on hydrated structure. To obtain detailed information on local environment of water change in different eutectic configurations, the 1H nuclear magnetic resonance spectrum (1H NMR) was conducted (Figure 2a). The 1H resonance of O−H in water is dislodged by high asymmetry eutectic coordination and experiences low‐field shift from 4.81 ppm in Aqua to 4.86 ppm in HAE, which suggests the redistributed electronic environment of 1H nuclei and weakened shielding effect of protons with lower electron density. Additionally, a downfield shift is recorded for the 1H peak along with the increased asymmetric degree of eutectic ligands, confirming the gradually enhanced breakdown of intrinsic bulk H2O‐H2O hydrogen bond network. The 1H resonance peaks of H2O show lower chemical shifts as the eutectic MA ratio changes in HAE‐1, HAE‐6, attributing to the enhanced dehydration in the formed zinc‐eutectic complex cluster (Figure S11). Of note, the smallest half‐peak width is obtained in HAE, indicating that the eutectic system composition with appropriate molar ratios is the most ordered and homogeneous. These observations are further reinforced by the Raman spectroscopy analysis between 3200 cm−1 and 4200 cm−1. As seen in Figure 2b, the broad Raman band of O−H stretching vibration modes of water molecules can be typically deconvoluted into three Gaussian components related to different hydrogen bond (HB) environments at around 3350 cm−1, 3550 cm−1, 3750 cm−1, corresponding to strong HB, weak HB and non‐HB, respectively. Of note, the broad peak shifts to a lower frequency with a narrower shape in eutectic mixtures, associating with the replacement of Zn2+‐H2O coordination with Zn2+‐eutectic (Figure S12). For HAE, the calculated percentage rearrangement of fitted area in peak unveils an apparently decreased occupation content of strong hydrogen bond water, whereas the non hydrogen bond one at higher Raman shifts magnifies a lot, which is assigned to the more severe disruption of hydrogen bond network and rearrangement of immobilized water (Figure 2c). It is assumed that such high level of mutual frustration for hydrogen bonding network of water molecules and eutectic ligands‐water pairs in HAE guarantees a frustrated and disordered coordination structure. [33]
Figure 2.
a) 1H NMR chemical shift of H2O in as‐designed electrolytes. b) The fitted Raman spectra from 3200 cm−1 to 4200 cm−1. c) Percentage of fitted area for strong hydrogen‐bond water and non hydrogen bond water. d) 3D snapshot of simulated HAE. e) Radial distribution functions of Zn‐H2O pairs. f) The simulated MSD versus time curves for hydrogen bond. g) in situ FTIR spectra of eutectic structure formation of HAE. h) Arrhenius plot of ionic conductivity and activation energy for different electrolytes.
To complement the Raman spectra testing, molecular dynamics (MD) simulations is conducted to intuitively describe the Zn2+ eutectic frameworks at different frustrated states. [34] The real‐time snapshot reveals that water, eutectic MA, and anion are engaged in the eutectic region via Zn−O interactions, verifying the transformation from mainly hydrous Zn2+ bound sheath to cation‐anion‐ligand‐water clusters (Figure 2d, Figure S13, Figure S14). To resolve the local structure of water in the modeling, the coordination numbers (dashed lines) and radial distribution function (RDF, solid lines) are employed. As displayed in Figure 2e, the sharp peak corresponding to the Zn2+‐O (H2O) is identified at the coordination distances of 2.4 Å. It suggests that the number of 5.8 H2O molecules involved in Aqua with aqueous solvation easily decreases in eutectic state networks with demonstrated order of LAE (3.6) > MAE (3.4) > HAE (3.1), indicative of a stronger destroying ability to the original water structure with asymmetric MA. The delocalizing charge densities between hydrogen bond accepter (C=O) and donor (N−H) on eutectic ligands contribute to hydrogen bonding interaction formation. In this regard, MA with more asymmetric structure compared to UA and AA can induce relatively weak interaction with water molecules, leading to disordered distribution of hydrogen bonding in eutectic domains. Meanwhile, an apparent change is also presented in hydrogen bonding distribution curves derived from mean square displacement (MSD), in which HAE exhibits the smallest number among different electrolytes and varies dramatically compared to Aqua, which reveals enhanced disorganization and breakage of pristine hydrogen bonds, just resembling the sequence result in RDF (Figure 2f). The variation of peak location and intensity in a series of in situ FTIR could be observed more visually from the corresponding contour plot. As shown in Figure 2g, the intensity of broad O−H stretching peak from region 3200–3700 cm−1 originally attributed to water in the tracked structure evolution becomes weak and center at 3400 cm−1, depicting that the eutectic ligands gradually participate in coordination structure and water significantly de‐cooperate from continuously original hydrogen bonding frameworks. [35] This phenomenon is more obvious in HAE compared to LAE, which demonstrates an obvious difference in degree of disorder characterized by a maximum disruption of hydrogen bonding network and inclusion of water from eutectic coordination domains (Figure S15). Based on Vogel‐Fulcher‐Tammann equation and Arrhenius fitting of conductivities measured at varying temperatures, the notable disparity is observed in the ionic mobility as high value is recorded along with increased eutectic frustration degree in HAE (Figure 2h). Meanwhile, a smaller driving force is required in HAE (20.8 kJ/mol) for Zn2+ plating compared to energy barrier in LAE (29.2 kJ/mol) and MAE (24.4 kJ/mol). The high structure order of local eutectic clusters in LAE may introduce coordinated aggregates, as evidenced by the sudden drop of ionic conductivities at −20 °C (Figure S16). More explicitly, the relatively lower probability of producing tetrahedral ordered hydrogen bonding network or abduction of intermolecular interactions in HAE compared to LAE and MAE can lead to more disordered distribution of water molecules and facilitated ion migration, in combination with the lower activity barrier in eutectic electrolyte, resulting in a high degree of ion motion even at low temperature.
Coordination Chemistry of Cation, Ligand and Anion
Next, we consider the term (ii) how the coordination environment of eutectic ligands and anions is affected by high frustration and asymmetry. In FTIR analysis, an apparent change is illustrated in stretching mode of functional group C=O, which is closely associated with accessible metal‐oxygen coordination between C=O group (Lewis base) and Zn2+ (Lewis acid) (Figure S17). The peaks gradually get narrowed and redshift from 1662 cm−1 (LAE) to 1628 cm−1 (HAE), indicating that charge distribution on the C=O group is altered and the weakened intermolecular interactions between HAE and Zn2+ in coordination shell. An inductive effect on the C=O band of MA can be induced molecular asymmetry and reduction in transferred negative charge on oxygen atom is enabled, which consequently weakens the interaction between Zn2+ with MA. This is also demonstrated by the 1H NMR spectra, where the 1H chemical shift of CH3‐C=O for HAE‐x exhibits a clear downfield shift with increasing ratio of MA due to the transfer effect of O atom from carbonyl participating in Zn2+ coordination (Figure 3a). Concomitantly, the 1H resonance of CH3‐NH at 2.71 ppm undergoes the incremental intensity from HAE‐1 to HAE‐6, which demonstrates gradually enhanced participation of MA in coordination shell (Figure S18a). Of note, similar behavior can be inferred from the 1H peak of CH3‐C=O in different ligands, as it experiences a slight upfield shift in HAE (1.941 ppm) compared to MAE (1.967 ppm), caused by the weakened de‐shielding effect and decreased binding (Figure S18b).
Figure 3.
a) NMR spectra of 1H peak of CH3‐C=O in MA from 1.9 ppm to 2 ppm. b) Raman spectra of Aqua, LAE. MAE and HAE. c) Solvate species distribution. d) Radial distribution functions of Zn‐ligand pairs. e) Radial distribution functions of Zn‐anion pairs. f) Zn K‐edge XANES spectra. g) Fourier‐transform EXAFS curves of LAE, MAE and HAE electrolytes. h) WT‐EXAFS of LAE, MAE and HAE electrolytes.
The Raman spectra were further characterized to assess the chemical environment of ligands (Figure 3b). Compared with the band at 898 cm−1 associated with vibration of HN−C=O group in LAE, the peak region of HAE (871 cm−1) and MAE (883 cm−1) is apparently decreased with redshift mode. This should be an indicator of the mitigated electric polarizability with lowered peak intensity, consistent with the notion of changed charge distribution on the amide group due to frustrated coordination and asymmetry of ligand. [36] The band information at 930 cm−1 assigned to Zn−O vibration in [Zn(ClO4)]+ can also shed light on the details of Zn2+‐ClO4 − interaction. Three modules can be divided from the characteristic bands located at around 928, 934, 940 cm−1, corresponding to free anions (FA), solvent separated ion pairs (SSIP), and contact ion pairs (CIP), respectively [37] (Figure S19). The percentages of each component based on the area for fitted peaks are summarized in Figure 3c. The cation‐anion binding at high disordered and asymmetric eutectic states is sharply heightened to generate ion clusters in MA. This is in accordance with the observation in FTIR results, as these sharp and separated bands referring to NH2 located in the range of 3200–3600 cm−1 in LAE and MAE gradually weakened as the frustration and asymmetry eutectic effect is strengthened in HAE (Figure S20). Less symmetric molecular geometry and more coordination disorder may help ClO4 − anion to release from electron‐donor sites of NH and decrease strong salt‐ligand association, leading to more involvement of anions in the primary coordination structure in HAE (Figure S21).
MSD simulation and RDF results are further discussed here to extract detailed information on ligand and anion coordination. As illustrated in Figure 3d, the average coordination number with a tendency of Zn2+‐O (UA) > Zn2+‐O (AA) > Zn2+‐O (MA) pair describes a gradually decreased coordination intensities along the increased frustration and asymmetry in eutectic ligands, indicating the inductive effect on C=O with reduced negative charge transfer in HAE. Meanwhile, HAE shows particularly obvious ClO4 − concentration gap with higher coordination number in the surrounding zone of Zn2+ compared to LAE and MAE, which suggests more anions engagement in the eutectic region, corroborating the results obtained from Raman analysis (Figure 3e). Namely, owing to the enhanced asymmetry of ligand and increased frustration of eutectic coordination structure, the cation‐ligand coupling in eutectic networks gets less locally bounded at the molecular level, with relatively intimate [Zn(ClO4)]+ cluster resides in shell, thus preserving the low viscosity, and achieving fast ion diffusion.
The structure of high frustration eutectic electrolyte enabled by HAE was also investigated by X‐ray adsorption fine structure (XAFS) technique. As illustrated in X‐ray adsorption near‐edge structure (XANES) spectra, the edge energy is sorted with sequence of Zn<ZnO<LAE<MAE<HAE<Aqua electrolytes as indicator of the effective charge on Zn (Figure 3f). The as‐prepared electrolytes with unique local structures all exhibit positive shift of adsorption edge position compared to metal Zn foil and ZnO, signifying the higher Zn valence state. Apparently, compared to LAE and MAE, Zn spectrum of HAE exhibits slightly risen average edge energy, which is possibly attributed to lower electron donating ability of C=O on asymmetric MA when coordinating Zn2+, contributing to less electrons around center of Zn2+. It is reasonable to speculate that introducing high frustration eutectic electrolyte may cause asymmetric MA ligand de‐cooperating from the tight interaction with Zn2+, thus providing reduced charge transfer from Zn2+ in coordination shell. The Fourier transformation of X‐ray absorption fine structure (EXAFS) with oscillation in the k‐space is fitted to analyze the local structures precisely (Figure S22). Longer Zn−O bond radial distances at strong signal of 1.56 Å can be observed in HAE, proving less enhanced interaction and loose coordination structure (Figure 3g). Additionally, the wavelet transformed (WT) EXAFS was adopted to analyze the atomic configuration (Figure 3h). The distance of Zn−O coordination is fitted as 1.58 Å in HAE electrolyte and is slightly longer compared to MAE (1.50 Å) and LAE (1.46 Å), which depicts longer bond length of Zn−O and weakened binding, validating the relatively reduced cation‐ligand correlated motion after establishing high frustration eutectics.
Attainment of Homogeneous and Reversible Zinc Plating/Stripping
Subsequently, the interfacial compatibility is evidenced by the plating/stripping kinetics of zinc electrodes and nucleation barriers. The electrochemical performances of HAE and three reference electrolytes were evaluated under Coulombic efficiency measurements over a wide range of current densities in Zn//Cu cells, in which HAE exhibits stable deposition/dissolution process, suggesting the potential for hindered side reactions (Figure 4a, Figure S23). Aqua electrolyte displays irreversible polarization at 5 mA cm−2, revealing tortuous migration path for Zn2+ and by‐product formation in aqueous environment. Of note, lower average CE accompanied by larger voltage hysteresis can be found for LAE and MEA. This phenomenon is especially demonstrated by smaller nucleation energy barrier in HAE, where the required overpotential of Zn is well retained to 41 mV owing to high hydrophilicity and improved ion transfer, indicating the intrinsically high frustration coordination concept toward highly durable Zn (Figure S24). As shown in Figure 4b and Figure S25, HAE‐based cells illustrate stable operation and average CE of 98.7 %, 99.4 % at 2 mA cm−2, 1 mA cm−2 respectively, which verifies the suppression ability of parasitic reactions at interface, vastly differing from the apparent oscillation ones in LAE and MAE electrolytes. To further validate the benefit of disordered coordination derived from asymmetric eutectic ligand in electrolytes, cycling stability of Zn//Zn symmetrical cells on long‐time scale is also tested (Figure 4c). The battery operation with less than 100 h cycles is enabled by Aqua at 1 mA cm−2, followed by a voltage surge and sudden short circuit. Contrariwise, excellent cycling performance with long‐term lifespan are realized in HAE electrolyte for 5000 h, which is also confirmed by the stable voltage profile at 4 mA cm−2 (Figure S26). Moreover, the elevated Zn reversibility can be discerned at high plating capacity of 5 mAh cm−2, with an uplifted cycling life over 200 h, almost tripled than that of LAE and MAE. This superiority of HAE has not been compromised even in rate performance with a stepwise increasing current density from 0.5 to 5 mA cm−2 (Figure 4d). Notably, Aqua and LAE all deliver progressively short voltage profiles, and experience sudden break down at low current densities, which can be ascribed to unidirectional growth of dendrite and corrosion. Regarding the various content of MA, HAE displays increased cycling numbers and a relatively reduced voltage hysteresis without augmented fluctuation compared to HAE‐1, HAE‐6 (Figure S27). It further identifies the upgraded Zn plating/stripping behavior that higher content of MA cannot alter and maintain eutectic coordination efficiently, while lower percentage of MA may lead to sluggish kinetics due to the high zinc ion concentration. Furthermore, a quantitative comparison of the key parameters for eutectic electrolytes is summarized in Figure 4e, highlighting the preponderance of the configurational disorder strategy[ 7 , 38 , 39 , 40 , 41 , 42 , 43 ] (Table S1). Encouragingly, the Zn electrochemistry with other commonly used, symmetric eutectic ligands is also investigated (Figure S28). In comparison, the prepared eutectic electrolytes from sulfolane, methylsulfonylmethane, and ethylene glycol all exhibit observably decreased battery lifetime with corresponding lower CE, which substantiates the adaptability of frustrated coordination concept.
Figure 4.
a) Rate performance of asymmetric cells with Aqua and HAE. b) Cycling performance of Zn//Cu asymmetric cell in different electrolytes. c) Zinc plating/stripping in symmetric cells at 1 mA cm−2. d) Rate performance of symmetric cells with Aqua, LAE, MAE and HAE. e) Comparison of reported eutectic electrolytes works regarding CE and stability at 1 mA cm−2. f) Chronoamperograms of Zn foils measured under −150 mV in different electrolytes. g) Tafel plots of the Zn foils tested in Aqua, LAE, MAE and HAE. h) 3D Raman images of cycled Zn in Aqua and HAE. i) The 3D confocal microscope images of cycled Zn in different electrolytes with skewness Rsk. j) SEM images of deposited Cu foil surface after electrochemical cycling. k) Cross‐sectional digital images of Zn deposition morphology evolution in optical microscope.
Then, the influence Zn2+ deposition kinetic and regulation of electrochemical responses in frustration‐driven eutectic electrolyte were studied. Chronoamperometry (CA) analysis was conducted to detect adsorption change of electroactive ions at interface in current‐time profiles (Figure 4f). The current response in HAE reaches its steady state soon in most constrained 2D diffusion and stable 3D diffusion way, implying the saturation of all nucleation sites. In contrast, a constantly rampant 2D diffusion pattern process is governed in Aqua during Zn plating, signifying the progressively rough propagation of nucleation sites. Additionally, the CV tests using half cells also prove the effect of HAE on the initial nucleation behavior, in which the reduced nucleation overpotential is detected compared to LAE and MAE, confirming uniform formation of Zn nucleation sites and enhanced adsorption (Figure S29, Figure S30). Furthermore, the improved corrosion resistance of HAE is verified in Tafel curves as larger corrosion potential and lower corrosion current density are displayed in Figure 4g. Meanwhile, the widened electrochemical window of HAE with higher H2/O2 evolution potentials than Aqua electrolyte manifests low kinetics of water splitting and restrained proton transfer (Figure S31). As a result, HAE endowed with controllable nucleation and limited side reactions is in favor of the reversible deposition upon cycling.
To better elucidate Zn deposition behavior under the influence of frustrated coordination in eutectic electrolyte, the morphology evolution of cycled Zn was examined in 3D Raman images (Figure S32). As shown in Figure 4h, large amount of dislocations is introduced on Zn surface after cycling in Aqua, which may lead to pulverization and fracture of electrodes. On the contrary, the homogeneous interface with gradually reduced granular deposition can be attained as asymmetric ligand‐involved eutectic HAE is applied (Figure S33). 3D confocal laser scanning microscopy (CLSM) images were then detected to monitor the morphology change (Figure 4i). A relatively compact‐packed morphology appears on Zn surface in HAE with lower skewness, which suggests tense and uniform Zn deposit layer without obvious dendrite after battery operation, certifying the elevated Zn reversibility. The four samples show a crucial difference in morphology observed in SEM images (Figure 4j). It is noteworthy that very rough Zn electroplating proceeds along interface with voids or cracks in Aqua. Oppositely, the surface topography can evidently become dense and smooth as the asymmetric level increases from LAE to HAE (Figure S34). The above result coincides with the undertaken optical microscopy observation, in which protuberances can be clearly detected in Aqua, while Zn plating remains flat and compact in HAE without many bulges (Figure 4k, Figure S35).
Robust SEI Contributes to Uniform Zinc Deposition
To visually show the different effects on interfacial stability caused by high‐frustration driven eutectic constituents, the composition of driven interphase formed on zinc anode is first characterized by X‐ray diffraction (XRD). As depicted in Figure 5a, the characteristic peaks corresponding to Zn5(OH)8Cl2 ⋅ H2O (ZnClOH) and ZnO on the cycled zinc surface are observed, as evidenced by a distinct peak at 10.6° and 34.7°, respectively. The intensity peak of ZnClOH gradually enhances when involving high asymmetric eutectic as seen in HAE, whereas it is the opposite case in signal of ZnO as decrement is found. The reversal suggests diverse interphase architecture raised in different eutectic systems. It also appears that there is a considerably higher intensity ratio of I(002)/I(100) as the asymmetric level increases, indicating the interior of grains and phase recognition in the presence of HAE [44] (Figure 5b). The XRD analysis also confirms the corrosion inhibition capability, as the severe by‐product formation after immersion in Aqua and LAE becomes significant. Conversely, the surface in HAE displays minimal precipitation peaks (Figure S36). The propensity of Zn anodes to form SEI under prolonged aging is further evaluated via in‐depth X‐ray photoelectron spectroscopy (XPS) (Figure 5c). As the sputtering proceeds, the strong C 1s signals involving (C−C/C−H, C−O/C−Cl) and (O‐C=O) components due to eutectic decomposition decrease significantly. The coverage of organic species in eutectic electrolytes suggests that eutectic MA may experience further electrochemical reduction and nucleophilic attack (Figure S37). The statistical results are also verified by DFT calculation of the LUMO variation (Figure S38). The LUMO energy level of Zn2+‐MA is lower than that of Zn2+‐UA and Zn2+‐AA. Collectively, the coordinated asymmetric ligand will preferentially tend to accept electrons and go through nucleophilic decomposition into the corresponding precipitated organic species. Figure 5d shows that Cl−Zn‐rich, Cl−C‐inclusive signal gradually dominates along with sputtering. The content of Cl‐related products originated from decomposition of Zn(ClO4)2 salt and eutectic ligand remain relatively high in HAE, implying considerably rich inorganic species regardless of the outer or inner layer. While obvious contrast can be achieved in Cl‐involved species for different eutectic variability plots, where only subtle phase transitions of Cl−Zn are observed in inner region of Aqua (Figure S39). Specifically, Zn, C, Cl, and O element arrangement distribution for different electrolytes are provided in Figure 5e, confirming enhanced anion participation in coordination shell regulated by asymmetric ligand of HAE.
Figure 5.
a) X‐ray diffraction spectra of cycled Zn in various electrolytes. b) Comparison of intensity ratio of I(ZnClOH)/I(ZnO) and I(002)/I(100). c) C 1s XPS spectra of the deposition formed on Zn anode. d) Cl 2p XPS spectra of the deposition formed on Zn anode. e) Atomic contents derived from the XPS spectra of deposition in various electrolytes. f) 3D views of CN−, C−, ZnO−, and ZnCl− distributions in the formed interphase in HAE. g) Depth profiling of CN−, C−, ZnO−, and ZnCl− signals collected from TOF‐SIMS. h) SEM images and corresponding EDX mapping of cycled Zn in HAE. i) TEM images of interphase on zinc electrode after cycling in HAE and Aqua. j‐l) TEM images of the selected area. m) Possible reaction Schemes of MA under electrochemical decomposition/nucleophilic attack and formation of ZnClOH. n) Schematic illustration about the underlying mechanism.
TOF‐SIMS is also performed to resolve the relative 3D architecture and chemical components (Figure 5f). It can be visibly found that the distribution of these two species fragments (C−, CN−) displays inhomogeneous features and becomes thinner in deeper region. The selected fragments ZnO−, ZnCl− are dispersive and spread across the 3D structure. This visual representation demonstrates the distinct differences in the distribution and coverage of SEI components in HAE (Figure S40). Based on the corresponding depth profile curves, the signal intensity of elemental C− and CN− from eutectic decomposition sharply decrease, corroborating the results obtained from XPS (Figure 5g). While ZnO−, ZnCl− tend to dominate the whole SEI with increasing sputtering time, hinting at the accumulation of multiple inorganic chemical constituents in the layer. Furthermore, the morphology feature and construction of interphase on Zn anode in different electrolytes were investigated using focused ion beam‐scanning electron microscope (FIB‐SEM) and transmission electron microscope (TEM). The EDX mapping delineates the uniform dispersion of Zn, Cl, and O of zinc deposition layer in HAE (Figure 5h). As revealed in Figure S41 and Figure S42, the milled lamella displays a distinct surface layer between the top of main Zn substrate and dense deposition in HAE. However, there is no apparent SEI generated in Aqua with only localized bulky zinc deposition morphology. The zinc ion diffusion and deposition will be difficult in this tightly packed and rough deposit interphase. As displayed in Figure 5i, Figure S43 and Figure S44, the multi‐component nature of SEI is demonstrated by the observed amorphous and crystalized region with thickness of around 120 nm. A mixture of crystalline lattice fringes of ZnO (100), Zn5(OH)8Cl2 ⋅ H2O (101) are identified in the SEI, which may possess high interfacial energy to suppress the detrimental side reactions [45] (Figure 5j–5l). Compared to HAE, the poly‐inorganic composite aggregates and larger crystal lattice area can be detected with numerous non‐uniform protrusions in Aqua, which may irreversibly block ion transport and cause accumulation of by products in the stripping/plating processes.
Based on the above analysis, the possible reaction pathways for electrochemical decomposition and nucleophilic attack are illustrated below (Figure 5m). The zinc salt decomposition and anion dissociation‐reduction dominate the contribution of products′ reaction due to the preferential energy, leading to richly deposited inorganic species in layer. Then protons can attack carbonyl group of MA to produce an intermediate. Then anions will neutralize it in HAE with higher nucleophilic reactivity by nucleophilic attack. The corresponding products would further react to generate less soluble polymer precipitation with various fragments to act as shield toward structural aberrations. Specifically, the outermost part of SEI in HAE is constructed by organic products, which provide incorporation, flexibility and ion transfer properties. The dispersive abundant distribution of inorganic compounds exists throughout the interphase, acting as a passivation barrier and preventing the dendrite and accommodate the volume changes that occur within the interfacial structure.[ 46 , 47 ] All these contribute to the formation of a uniform and compact foundation of SEI in HAE. It is stated that the well‐organized and ion‐conductive SEI consisting of asymmetric ligand‐derived organic moieties and inorganic‐rich phases can guarantee highly reversible Zn2+ (Figure 5n). This combination of a flexible outer dense organic layer and an indurated inorganic moieties spreading out the whole matrix is crucial for maintaining the stability and functionality of SEI in HAE (Figure S45).
Practical Application to Zinc‐ion Batteries
Finally, the feasibility of frustrated coordination concept was exemplified in assembled zinc full cells coupling with VO2 as active cathode material (Figure S46). The good overlap of paired redox peaks at around 0.91 V/0.55 V in reduction cycle and 0.71 V/1.09 V in oxidation cycle for CV curves in different electrolytes‐based cells illustrate the reversible ion insertion/extraction reactions on VO2 cathode (Figure S47, Figure S48). Since V‐based material features high charge storage capability involving surface‐contribution advantages, it can retain fast ion kinetics when storing/releasing ions. Next, the rate performances of HAE, MAE, LAE at different current densities from 0.1 A g−1 to 3 A g−1 were investigated (Figure S49). An initial enhanced capacity of 463 mAh g−1 at 0.1 A g−1, and high capacity yield of 181 mAh g−1 at 3 A g−1 in HAE can be obtained. Although acceptable charge storage behavior is observed for low current cycles between Aqua and HAE, the difference in delivered capacity output gradually appears as the reaction proceeds to high current density. It can be assigned to protection derived from reduced free water activity and SEI formation against side reactions when fast intercalation/de‐intercalation reaction occurs. Of note, a more noticeable specific capacity decay drop of LAE is illustrated with increasing rate, suggesting that limited ionic mobility and mass transfer will reduce the virtue of eutectic mixture. Accordingly, as displayed in Figure 6a, promising capacity are achieved in HAE compared with LAE and MAE, which is associated with favorable eutectic affinity. Moreover, the charging and discharging curves at various current densities manifest a smaller voltage gap in HAE, which recrudesces at different current rates, corroborating the results obtained from CV tests and demonstrating fast rate capability (Figure S50). The advantages of frustrated coordination eutectic systems are also illustrated by the self‐discharge behavior and long‐term cycling performance (Figure S51). The capacity of full cells in Aqua rapidly declines, which can be evidenced by voltage polarization profiles (Figure 6b). On the contrary, the system with HAE still sustains high‐capacity retention as the cycling number increases. From the application viewpoint, cell configuration with low N/P ratios (limited anode and high cathode loading) is essential to realizing practical energy storage. In this case, the longevity benefited from the stabilization of Zn electrode can also be evident from the performance applying N/P ratio of 5.4 and 30 μm thick commercial zinc foil. As shown in Figure 6c, an 82 % capacity retention of the initial capacity is attained at 1 A g−1 in HAE after 3000 cycles with near‐unity CE (Table S2). Although the initial specific capacity of other reference electrolytes is not compromised, the fast capacity fading rate is hard to avoid due to the inferior interfacial kinetic barrier in the vicinity of Zn surface. The stable cycling of full cells with HAE relies on maintaining a favorable interface chemistry, ensuring good interfacial contact. Thus, Nyquist plot and distribution of relaxation time (DRT) were conducted to quantify the resistance upon charge and discharge cycles (Figure 6d, Figure 6e). The DRT analysis revealed that the interface impedance of HAE at medium frequencies of −2 to −1 associated with the double‐layer relaxation, the passivation layers, and the charge‐transfer process were relatively lower than LAE. [48] Meanwhile, the lower diffusion impedance of HAE also manifested, indicating that the promoted ion transport and uniform deposition in HAE are beneficial for reducing interfacial resistance and decreasing passivation during cycling. On the basis of previous analysis, the cycled zinc electrode using Aqua may exhibit unfavorable plating/stripping results caused by “dead” deposition and by products. In light of these, the post‐observation on the cycled Zn in full cells was further probed in CLSM images (Figure 6f). The zinc surface keeps a homogeneous and dense appearance without fragmentation in HAE, while harsh island‐like Zn with increased internal spaces can be observed in Aqua.
Figure 6.
a) Rate performance of Zn//VO2 batteries in different electrolytes. b) The charge/discharge voltage profiles of full cells in Aqua and HAE at 2 A g−1. c) Long‐term performance of Zn//VO2 cells by using different electrolytes at current density of 1 A g−1 with N/P of 5.4. Nyquist plots from galvanostatic EIS and corresponding DRT transition results in d) HAE and e) LAE. f) 2D confocal microscope images of Zn after cycled in full cells. g) Long‐term cycle performance of Zn//VO2 battery at −20 °C at 0.5 A g−1. h) Radar plots: the properties comparison of as‐designed electrolytes. i) Voltage polarization comparison in Zn//MnO2 batteries and LED light image. j) The voltage‐time curves of pouch cell using HAE and corresponding open circuit images under different testing condition.
Of note, eutectic HAE‐induced coordination structure will compete with the ordered water‐ice network and disrupt the distribution of locally ordered tetrahedral units during the phase transition process. Additionally, owing to the disordered interactions in HAE system with coordination frustration, the free energy may experience a decrement and significant shifts of freezing point to low temperature, in favor of the promoted interface kinetics and anti‐freezing behavior. [49] As displayed in Figure S52, HAE shows good compatibility with zinc electrode during cycling at 0 °C, in which the CE remains at 98.5 % even after 800 cycles due to reconstruction of disordered eutectic‐water complexes. To demonstrate the utility of eutectic electrolyte at sub‐zero temperature, the battery with HAE at −20 °C exhibits average capacity of 84 mAh g−1 within 1000 cycles, much higher comparing with fast capacity decay and ice nucleation in Aqua (Figure 6g, Figure S53). Consequently, a critical Radar plot describing favorable electroplating factors is discussed including competitive cycling performances, reasonable cost, and safe properties, suggesting the reliability of asymmetric ligand‐involved eutectic design (Figure 6h). The practicality of high frustration eutectic electrolyte is also demonstrated in full cells using MnO2 as cathode material (Figure S54, Figure S55). As illustrated in Figure 6i, the voltage gap between charging and discharging in HAE decreases and stays at relatively low value as the current rates increase. The fabricated cell can also operate in series to successfully power LED sets. Moreover, considering the performance evolution with soft packing, HAE electrolytes were assembled into 4*5 cm2 pouch cell configuration with VO2 as cathode as seen in Figure 6j. Except for a slight decrease in open circuit voltage, the stable profit recurs even under harsh working conditions such as bending to 90°, piercing, or partial cutting of pouch cell. The durable operation of pouch cell is substantiated by average capacity of 155 mAh g−1 and high CE after 200 cycles without leaking and swelling, which states a potential future application scenario (Figure S56, Table S3).
Conclusion
In this contribution, we presented a unique frustrated eutectic electrolyte structure driven by asymmetric ligand coordination. Through multiple theoretical and experimental endeavors, the disordered characteristics of electrolyte and improved stability of electrodes are investigated due to the combination of several changes: (1) MA as high asymmetry eutectic ligand increases the configurational frustration and reduces the orderly self‐association in HAE eutectic electrolyte, thus promoting stability and diffusivity while maintaining eutectic property compared to MAE and LAE; (2) the inductive effect on the carbonyl group of MA ligand induced by molecular asymmetry lowers the transferred negative charge and weakens interaction with Zn2+, resulting in minimized engagement in coordinating; (3) the structural disorder with increased frustration effect in HAE helps to disrupt the local distribution of hydrogen bonding network; (4) the stronger tendency of cation‐anion binding in asymmetric ligand‐involved shell of HAE produce a dense and inorganic‐rich hybrid SEI. These merits enable compact zinc plating and improved electrochemical reversibility. Noticeably, high average CE of 99.4 % and durable Zn plating/stripping for 5000 h are obtained in HAE. The Zn//VO2 full cells with HAE also deliver excellent cycling performance under small N/P ratio and low‐temperature conditions. We believe that full recognition of asymmetric ligand‐coordinated structure with high configurational frustration will guide the development of better eutectic electrolyte systems for aqueous zinc batteries.
Conflict of Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
1.
Supporting information
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Supporting Information
Acknowledgments
This work was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), through the Discovery Grant Program (RGPIN‐2018‐06725) and the Discovery Accelerator Supplement Grant program (RGPAS‐2018‐522651), and by the New Frontiers in Research Fund‐Exploration program (NFRFE‐2019‐00488). Prof. X. Wang acknowledges the support from Canada Research Chair Program (CRC‐2022‐00059). This research was supported by funding from the Canada First Research Excellence Fund as part of the University of Alberta's Future Energy Systems research initiative (FES−T06‐Q03). The authors also acknowledge the support of NanoFAB in sample preparation and electron microscopy at the University of Alberta in Canada.
Deng W., Deng Z., Zhang X., Chen Y., Feng R., Li G., Wang X., Angew. Chem. Int. Ed. 2025, 64, e202416482. 10.1002/anie.202416482
Contributor Information
Prof. Ge Li, Email: ge.li@ualberta.ca.
Prof. Xiaolei Wang, Email: xiaolei.wang@ualberta.ca.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Wang R., Yao M., Yang M., Zhu J., Chen J., Niu Z., Proceedings of the National Academy of Sciences 2023, 120, e2221980120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Qiu M., Sun P., Han K., Pang Z., Du J., Li J., Chen J., Wang Z. L., Mai W., Nat. Commun 2023, 14, 601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Xiong P., Lin C., Wei Y., Kim J.-H., Jang G., Dai K., Zeng L., Huang S., Kwon S. J., Lee S.-Y., Park H. S., ACS Energy Lett. 2023, 8, 2718–2727. [Google Scholar]
- 4. Yang K., Fu H., Duan Y., Ma Z., Wang D., Li B., Park H. S., Ho D., ACS Energy Lett. 2024, 9, 209–217. [Google Scholar]
- 5. Deng W., Li G., Wang X., Adv. Funct. Mater. 2024, n/a, 2405012. [Google Scholar]
- 6. Ming F., Zhu Y., Huang G., Emwas A.-H., Liang H., Cui Y., Alshareef H. N., J. Am. Chem. Soc. 2022, 144, 7160–7170. [DOI] [PubMed] [Google Scholar]
- 7. Wang G., Fu H., Lu J., Huang S., Pei C., Min D., Zhang Q., Park H. S., Adv. Energy Mater. 2024, 14, 2303549. [Google Scholar]
- 8. Liu S., Mao J., Pang W. K., Vongsvivut J., Zeng X., Thomsen L., Wang Y., Liu J., Li D., Guo Z., Adv. Funct. Mater. 2021, 31, 2104281. [Google Scholar]
- 9. Zhao Y., Guo S., Chen M., Lu B., Zhang X., Liang S., Zhou J., Nat. Commun 2023, 14, 7080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Zheng Z., Zhong X., Zhang Q., Zhang M., Dai L., Xiao X., Xu J., Jiao M., Wang B., Li H., Jia Y., Mao R., Zhou G., Nat. Commun 2024, 15, 753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Li G., Zhao Z., Zhang S., Sun L., Li M., Yuwono J. A., Mao J., Hao J., Vongsvivut J., Xing L., Zhao C.-X., Guo Z., Nat. Commun 2023, 14, 6526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Tang L., Xu Y., Zhang W., Sui Y., Scida A., Tachibana S. R., Garaga M., Sandstrom S. K., Chiu N.-C., Stylianou K. C., Greenbaum S. G., Greaney P. A., Fang C., Ji X., Angew. Chem. Int. Ed. 2023, 62, e202307212. [DOI] [PubMed] [Google Scholar]
- 13. Dong D., Wang T., Sun Y., Fan J., Lu Y.-C., Nature Sustainability 2023, 6, 1474–1484. [Google Scholar]
- 14. Huang J., Zhong Y., Fu H., Zhao Y., Li S., Xie Y., Zhang H., Lu B., Chen L., Liang S., Zhou J., Adv. Mater. 2024, n/a, 2406257. [DOI] [PubMed] [Google Scholar]
- 15. Qu K., Lu X., Jiang N., Wang J., Tao Z., He G., Yang Q., Qiu J., ACS Energy Lett. 2024, 9, 1192–1209. [Google Scholar]
- 16. Li J., Lou Y., Zhou S., Chen Y., Zhao X., Azizi A., Lin S., Fu L., Han C., Su Z., Pan A., Angew. Chem. Int. Ed. 2024, n/a, e202406906. [DOI] [PubMed] [Google Scholar]
- 17. Geng L., Meng J., Wang X., Han C., Han K., Xiao Z., Huang M., Xu P., Zhang L., Zhou L., Mai L., Angew. Chem. Int. Ed. 2022, 61, e202206717. [DOI] [PubMed] [Google Scholar]
- 18. Xia Y., Zhou P., Kong X., Tian J., Zhang W., Yan S., Hou W.-h., Zhou H.-Y., Dong H., Chen X., Wang P., Xu Z., Wan L., Wang B., Liu K., Nat. Energy 2023, 8, 934–945. [Google Scholar]
- 19. Chen S., Li S., Ma L., Ying Y., Wu Z., Huang H., Zhi C., Angew. Chem. Int. Ed. 2024, 63, e202319125. [DOI] [PubMed] [Google Scholar]
- 20. Yang C., Xia J., Cui C., Pollard T. P., Vatamanu J., Faraone A., Dura J. A., Tyagi M., Kattan A., Thimsen E., Xu J., Song W., Hu E., Ji X., Hou S., Zhang X., Ding M. S., Hwang S., Su D., Ren Y., Yang X.-Q., Wang H., Borodin O., Wang C., Nature Sustainability 2023, 6, 325–335. [Google Scholar]
- 21. Zhao X., Fu Z., Zhang X., Wang X., Li B., Zhou D., Kang F., Energy Environ. Sci. 2024, 17, 2406–2430. [Google Scholar]
- 22. Wang Q., Zhao C., Yao Z., Wang J., Wu F., Kumar S. G. H., Ganapathy S., Eustace S., Bai X., Li B., Lu J., Wagemaker M., Adv. Mater. 2023, 35, 2210677. [DOI] [PubMed] [Google Scholar]
- 23. Wang Y., Li Z., Xie W., Zhang Q., Hao Z., Zheng C., Hou J., Lu Y., Yan Z., Zhao Q., Chen J., Angew. Chem. Int. Ed. 2024, 63, e202310905. [DOI] [PubMed] [Google Scholar]
- 24. Lin R., Ke C., Chen J., Liu S., Wang J., Joule 2022, 6, 399–417. [Google Scholar]
- 25. Piao Z., Wu X., Ren H.-R., Lu G., Gao R., Zhou G., Cheng H.-M., J. Am. Chem. Soc. 2023, 145, 24260–24271. [DOI] [PubMed] [Google Scholar]
- 26. Wu J., Liang Q., Yu X., Lü Q.-F., Ma L., Qin X., Chen G., Li B., Adv. Funct. Mater. 2021, 31 2011102. [Google Scholar]
- 27. Han M., Huang J., Xie X., Li T. C., Huang J., Liang S., Zhou J., Fan H. J., Adv. Funct. Mater. 2022, 32, 2110957. [Google Scholar]
- 28. Yang W., Du X., Zhao J., Chen Z., Li J., Xie J., Zhang Y., Cui Z., Kong Q., Zhao Z., Wang C., Zhang Q., Cui G., Joule 2020, 4, 1557–1574. [Google Scholar]
- 29. Su Y., Rong X., Li H., Huang X., Chen L., Liu B., Hu Y.-S., Adv. Mater. 2023, 35, 2209402. [DOI] [PubMed] [Google Scholar]
- 30. Gilbert A. S., Thermochim. Acta 2007, 452, 135–139. [Google Scholar]
- 31. Mackanic D. G., Yan X., Zhang Q., Matsuhisa N., Yu Z., Jiang Y., Manika T., Lopez J., Yan H., Liu K., Chen X., Cui Y., Bao Z., Nat. Commun 2019, 10, 5384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Wang H., Kim S. C., Rojas T., Zhu Y., Li Y., Ma L., Xu K., Ngo A. T., Cui Y., J. Am. Chem. Soc. 2021, 143, 2264–2271. [DOI] [PubMed] [Google Scholar]
- 33. Chen M., Runge T., Wang L., Li R., Feng J., Shu X.-L., Shi Q.-S., Carbohydr. Polym. 2018, 200, 115–121. [DOI] [PubMed] [Google Scholar]
- 34. Egorov A. V., Lyubartsev A. P., Laaksonen A., The Journal of Physical Chemistry B 2011, 115, 14572–14581. [DOI] [PubMed] [Google Scholar]
- 35. Qiu H., Du X., Zhao J., Wang Y., Ju J., Chen Z., Hu Z., Yan D., Zhou X., Cui G., Nat. Commun 2019, 10, 5374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Jiang H., Tang L., Fu Y., Wang S., Sandstrom S. K., Scida A. M., Li G., Hoang D., Hong J. J., Chiu N.-C., Stylianou K. C., Stickle W. F., Wang D., Li J., Greaney P. A., Fang C., Ji X., Nature Sustainability 2023, 6, 806–815. [Google Scholar]
- 37. Tian Z., Zou Y., Liu G., Wang Y., Yin J., Ming J., Alshareef H. N., Adv. Sci. 2022, 9, 2201207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Chen R., Zhang C., Li J., Du Z., Guo F., Zhang W., Dai Y., Zong W., Gao X., Zhu J., Zhao Y., Wang X., He G., Energy Environ. Sci. 2023, 16, 2540–2549. [Google Scholar]
- 39. Li W., Kong W., Liu W., Xu S., Zhu H., Liu S., Yu W., Wen Z., Energy Storage Mater. 2024, 65, 103103. [Google Scholar]
- 40. Zhong Y., Xie X., Zeng Z., Lu B., Chen G., Zhou J., Angew. Chem. Int. Ed. 2023, 62, e202310577. [DOI] [PubMed] [Google Scholar]
- 41. Wang S., Liu G., Wan W., Li X., Li J., Wang C., Adv. Mater. 2024, 36, 2306546. [DOI] [PubMed] [Google Scholar]
- 42. Wan J., Wang R., Liu Z., Zhang S., Hao J., Mao J., Li H., Chao D., Zhang L., Zhang C., Adv. Mater. 2024, 36, 2310623. [DOI] [PubMed] [Google Scholar]
- 43. Wang J., Qiu H., Zhang Q., Ge X., Zhao J., Wang J., Ma Y., Fan C., Wang X., Chen Z., Li G., Cui G., Energy Storage Mater. 2023, 58, 9–19. [Google Scholar]
- 44. Wang X., Ying Y., Li X., Chen S., Gao G., Huang H., Ma L., Energy Environ. Sci. 2023, 16, 4572–4583. [Google Scholar]
- 45. Zhang B., Wu X., Luo H., Yan H., Chen Y., Zhou S., Yin J., Zhang K., Liao H.-G., Wang Q., Zou Y., Qiao Y., Sun S.-G., J. Am. Chem. Soc. 2024, 146, 4557–4569. [DOI] [PubMed] [Google Scholar]
- 46. Wang W., Chen S., Liao X., Huang R., Wang F., Chen J., Wang Y., Wang F., Wang H., Nat. Commun 2023, 14, 5443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Niu C., Liu D., Lochala J. A., Anderson C. S., Cao X., Gross M. E., Xu W., Zhang J.-G., Whittingham M. S., Xiao J., Liu J., Nat. Energy 2021, 6, 723–732. [Google Scholar]
- 48. Yao L., Liu J., Zhang F., Wen B., Chi X., Liu Y., Nat. Commun 2024, 15, 6249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Wang D., Peng H., Zhang S., Liu H., Wang N., Yang J., Angew. Chem. Int. Ed. 2023, 62, e202315834. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Supporting Information
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.






