Quasi‐Solid Cathode Additive Enables Highly Reversible Four‐Electron I⁻/I⁰/I⁺ Conversion in Aqueous Zn‐I₂ Batteries

Abstract

Aqueous zinc‐iodine (Zn‐I₂) batteries with four‐electron (4e) I⁻/I⁰/I⁺ conversion (4eZIBs) offer high energy density but face both‐step I⁻/I⁰ and I⁰/I⁺ challenges, including the polyiodide shuttle effect, sluggish I⁰/I⁺ conversion kinetics, and severe I⁺ hydrolysis. To mitigate these issues, a quasi‐solid additive composed of 1‐butyl‐3‐methylimidazolium chloride (BMICl) and carbon nanotubes (CNTs) is introduced into the cathode. Specifically, by co‐grinding BMICl with CNTs, a homogeneous quasi‐solid additive is formed due to the π–π stacking interactions between CNTs and imidazole rings. This additive not only suppresses the shuttle effect by binding with polyiodides in the first‐step I⁻/I⁰ conversion, but also enhances I⁺ conversion kinetics by immobilizing Cl⁻ inside the electrode and curbs I⁺ hydrolysis through forming a BMI‐ICl₂ complex in the second‐step conversion. This innovative approach enables the 4eZIBs to achieve a near‐theoretical specific capacity of 418.9 mA h g⁻¹ at 0.5C, while maintaining a robust lifespan of over 600 cycles with a capacity retention of 93.4% at 1C. Moreover, pouch cells under a high areal capacity of 7.1 mA h cm⁻² for each side of the cathode demonstrate a high‐capacity retention of 95.8% after 150 cycles at 6.3 mA cm⁻² (≈0.5C).

A quasi‐solid BMICl/CNTs additive addresses dual challenges in four‐electron zinc‐iodine batteries (4eZIBs). It suppresses polyiodide shuttling via bonding in the I⁻/I⁰ step and enhances I⁺ kinetics while curbing hydrolysis via Cl⁻ immobilization and BMI‐ICl₂ complex formation in the I⁰/I⁺ step. This enables 418.9 mA h g⁻¹ capacity at 0.5C and 93.4% retention after 600 cycles at 1C.

1. Introduction

The increasing need for efficient and sustainable energy storage solutions has intensified the exploration of alternatives to traditional lithium‐ ion batteries.^{[ 1} , ² , ³ , ⁴ , ⁵ , ^{6 ]} Among various options, aqueous zinc batteries (AZBs), particularly aqueous zinc‐iodine batteries (ZIBs), have emerged as compelling candidates due to their inherent safety, low cost, environmental friendliness, and impressive energy and power densities.^{[ 5} , ⁷ , ⁸ , ^{9 ]} Traditional ZIBs operate on a two‐electron (2e) I⁻/I⁰ redox reaction, offering limited energy density compared to systems like Zn‐MnO₂.^{[ 10 ]} In response, 4eZIBs utilizing the I⁻/I⁰/I⁺ redox pathway have been developed to achieve higher redox potentials and double theoretical capacity.^{[ 11} , ¹² , ¹³ , ^{14 ]}

Differing from 2eZIBs, the practical application of 4eZIBs faces significant challenges during I⁻/I⁰ and I⁰/I⁺ conversions.^{[ 15 ]} For the first‐step I⁻/I⁰ conversion, a key issue arises from the generation of highly soluble polyiodides,^{[ 16 ]} such as I₃ ⁻ and I₅ ⁻, which trigger severe shuttle effects, resulting in low Coulombic efficiency (CE) and significant self‐discharge.^{[ 17} , ^{18 ]} Traditional electrolyte/electrode optimization strategies to confine polyiodides diffusion have been widely reported to mitigate the shuttle effects and suppress the relevant side reactions.^{[ 19} , ²⁰ , ^{21 ]} For the second‐step I⁰/I⁺ conversion, challenges include the slow conversion kinetics and the hydrolysis of I⁺ species (e.g., ICl or ICl₂ ⁻) during cycling.^{[ 22 ]} Currently, the activation of I⁺ conversion relies on electrolyte modifications, such as adding halide ions, like Cl⁻ ions, to stabilize the I⁺ ions.^{[ 23 ]} These strategies often involve incorporating high concentrations of Cl⁻ ions into the electrolyte to provide sufficient Cl⁻ ions and suppress the side reactions between water and I⁺ species by reducing the water activity.^{[ 24 ]} However, the high viscosity of the concentrated electrolyte and the sluggish diffusion of Cl⁻ from the electrolyte to the bulk of the electrode limit the conversion rate of I⁰/I⁺, especially under a high cathode (iodine) areal loading.^{[ 18} , ²³ , ^{25 ]} Although a few reports have focused on optimizing electrolytes to address both‐step conversion challenges, there are currently no electrode optimization strategies capable of overcoming weaknesses of both steps, especially through a one‐shot approach. Therefore, the innovative and one‐shot electrode solution is highly desirable to achieve advanced 4eZIBs with reversible I⁻/I⁰ and I⁰/I⁺ conversions.^{[ 26 ]}

Here, a quasi‐solid cathode additive strategy is reported by incorporating 1‐butyl‐3‐methylimidazolium chloride (BMICl) and carbon nanotubes (CNTs) directly into the cathode. The π–π stacking interactions between the CNTs and the imidazole rings in BMICl prompt the formation of a quasi‐solid additive inside the electrode structure. In the first‐step I⁻/I⁰ conversion, the electrode additive mitigates the polyiodide shuttle effect by combining the polyiodides and forming an insoluble precipitate. In the second‐step I⁰/I⁺ conversion, the additive not only curbs hydrolysis of I⁺ by binding ICl₂ ⁻ ions but also improves the conversion kinetics of I⁺ by providing Cl⁻ ions within the cathode. The 4eZIB coin‐cell configuration with the quasi‐solid additive achieves a high specific capacity of 418.9 mA h g⁻¹ (based on the mass of iodine) at a low current density of 0.5C, substantially higher than that of a NaCl‐added control group (317.5 mA h g⁻¹) in the same electrolyte (2 m ZnSO₄+0.5 m NaCl), while maintaining a robust cycle life of over 600 cycles with a capacity retention of 93.4% at 1C under a high cathode areal capacity over 3 mA h cm⁻². Furthermore, pouch cells incorporating quasi‐solid electrode additives exhibit high‐capacity retention of 95.8% after 150 cycles at a low current density of 6.3 mA cm⁻² (≈0.5C), even under a high areal capacity of 7.1 mA h cm⁻² for each side of the cathode. By addressing the fundamental challenges inherent in 4eZIBs and proposing a scalable, cost‐effective solution, this work contributes profoundly to the advancement of aqueous battery technologies, promising a new horizon in the quest for energy sustainability.

2. Results and Discussion

2.1. The Interaction between BMICl and CNTs

In previous studies, Cl⁻ ions were added to the electrolytes to activate the I⁰/I⁺ conversion, but these approaches are difficultly compatible with cathodes that feature high areal capacities due to the sluggish diffusion of Cl⁻ from the electrolyte to the bulk of the cathode, especially in thick ones. As the cathode thickness increases with areal capacity, the penetration of Cl⁻ from the electrolyte to the bulk of the cathode becomes more challenging, leading to incomplete I⁺ conversion. A possible solution involves integrating a portion of the Cl⁻ in the form of a chlorinated compound during the cathode fabrication process. This could potentially enhance the efficiency of Cl⁻ in activating the I⁺ conversion, thus increasing the battery capacity. Selecting an appropriate chlorinated compound involves two critical criteria: the compound must contain Cl⁻ or other halogen ions, with the ions in a dissociable state to participate in electrode reactions, and it should be insoluble in water. Based on former criteria, BMICl, a chlorinated ionic liquid, is identified as a candidate to provide Cl⁻. However, challenges emerged during experimental trials; BMICl is highly soluble in water and could rapidly dissolve into the electrolyte, undermining its effectiveness. Previous reported works confirmed that grinding imidazole‐based ionic liquids with CNTs could form a quasi‐solid substance due to the π–π stacking interactions between the CNTs and the imidazole rings.^{[ 27} , ^{28 ]} Similarly, the BMICl in this study can function with CNTs through π–π stacking interactions to generate the insoluble compound, significantly reducing the solubility of BMICl in the water‐based electrolyte (Figure S1, Supporting Information). The fabrication process of the BMICl electrode is illustrated in Figure 1a. Initially, CNTs and BMICl were added to a mortar. The mixture was ground for 30 min to form a quasi‐solid BMICl, as depicted in Figure S2 (Supporting Information). Subsequently, a mixture of I₂ and carbon (I₂@KB) was added, and grinded for an additional 30 min to ensure thorough mixing. Finally, a solution of polytetrafluoroethylene (PTFE) was incorporated into the mixture to prepare the BMICl‐based I₂ electrode. In comparison, the NaCl‐electrode was prepared following the same procedure as that for the BMICl‐electrode, with the substitution of NaCl in an equivalent molar ratio to BMICl (Details can be found in the experiment section). To determine the optimal BMICl content, a series of electrodes with varying molar ratios of BMICl to I (4:1, 2:1, and 1:1) were prepared and evaluated (Figure S3, Supporting Information). Both BMICl‐loaded electrodes with the molar ratio of 2:1 and 1:1 delivered high discharge capacities (≈400 mA h g⁻¹) at 1C with an I⁺ activation rate close to 50%, indicating effective activation of the I⁰/I⁺ redox couple. However, the ratio of 1:1 significantly increased the inactive mass of the electrode and reduced overall energy density. Therefore, the ratio of 2:1 was identified as the optimal BMICl content, providing a favourable balance between redox reversibility and practical energy density.

Figure 1.

Fabrication and interaction of the BMICl‐electrode. a) Schematic of the quasi‐solid BMICl formation process. b) Raman spectra showing interactions between carbon nanotubes (CNTs) and BMICl. c) UV–vis absorption spectra after immersing the electrode in the electrolyte for 2 h. d) Results of the sublimation tests for the electrodes.

The interaction between CNTs and BMICl was investigated using attenuated total reflectance‐fourier transform infrared spectroscopy (ATR‐FTIR) and Raman spectroscopy. As illustrated in Figure 1b, the Raman spectra reveal a blue shift in the peaks associated with CNTs from 1346 and 1585 cm⁻¹ to 1348 and 1592 cm⁻¹, respectively, indicating the interaction between CNTs and BMICl. As shown in Figure S4 and Table S1 (Supporting Information), the peak at 1183  cm⁻¹ is attributed to the in‐plane bending vibrations of C–H, while two new ATR‐FTIR peaks appearing at 1114 and 1084  cm⁻¹ after co‐grinding BMICl with CNTs are assigned to deformation modes of the BMICl molecules.^{[ 29} , ^{30 ]} The morphological feature of CNTs was further characterized by scanning electron microscopy (SEM). As shown in Figure S5 (Supporting Information), pristine CNTs aggregate into spherical bundles, whereas after mixing with BMICl, they become well dispersed and form a gel‐like interconnected structure, indicating the interaction between BMICl and CNT. To further probe the interfacial charge characteristics, Zeta potential measurements were performed in a 2 m ZnSO₄+0.5 m NaCl electrolyte (Figure S6, Supporting Information). Bare CNTs exhibited a slightly negative zeta potential of −0.146 mV, while the BMICl‐CNTs shifted to a positive value of +0.875 mV. This shift indicates that while the cationic BMI⁺ species are immobilized on the CNT surface via π–π interactions, the Cl⁻ ions remain dissociated in the electrolyte, leading to a positive surface charge. This positively charged interface may enhance the electrostatic attraction of polyiodide anions, thereby aiding their confinement during battery operation. In addition, contact angle measurements demonstrate a substantial enhancement in wettability after BMICl incorporation. As shown in Figure S7 (Supporting Information), the contact angle of the electrode surface decreases markedly upon BMICl functionalization, indicating improved hydrophilicity. This enhancement facilitates more effective electrolyte infiltration into the cathode matrix, ensuring uniform electrochemical reactions and reducing interfacial resistance under high cathode areal capacities. The solubility of BMICl in the electrolyte (2 m ZnSO₄+0.5 m NaCl) was assessed using UV–vis spectroscopy. The UV–vis spectroscopy was conducted in an operando configuration using a customized quartz cuvette cell (Figure S8, Supporting Information). The cathode was pressed onto a Ti mesh and dried. Both the Ti mesh‐supported cathode and Zn foil anode (each ≈0.8 cm × 6 cm) were vertically inserted into the cuvette filled with 2 m ZnSO₄ + 0.5 m NaCl electrolyte, allowing real‐time absorbance monitoring during cycling. When an equivalent amount of pure BMICl (without CNTs) was directly added into the electrolyte, a sharp increase in absorbance was observed within 32 min, indicating its rapid dissolution (Figure S9, Supporting Information). In contrast, the BMICl‐based electrode, prepared by co‐grinding BMICl with CNTs, showed no significant peak even after 2 h of immersion in the same electrolyte (Figure S10, Supporting Information). These results confirm that the π–π interactions between BMICl and CNTs effectively suppress BMICl leaching and ensure its stability in aqueous environments. Furthermore, post‐cycling energy‐dispersive spectroscopy (EDS) mapping of the BMICl‐based electrode, after thoroughly rinsing to remove residual salts, reveals strong N and Cl signals characteristic of BMICl, confirming its stable immobilization on CNTs during extended operation (Figure S11, Supporting Information).^{[ 31 ]} Moreover, no peak corresponding to I₂ was identified, suggesting that the dissolution of I₂ is suppressed by the quasi‐solid BMICl. In contrast, electrodes fabricated with NaCl exhibited a noticeable increase in peak intensity during the same rest period related to dissolved I₂ after immersion in the same electrolyte (Figure 1c; Figure S12, Supporting Information). Furthermore, the quasi‐solid BMICl within the electrode appears to mitigate the sublimation issue associated with I₂. After exposure to the atmosphere for 10 days, the NaCl‐electrode lost ≈7% of its mass due to the sublimation of I₂, whereas the BMICl‐electrode maintained over 99.5% of its mass, as shown in Figure 1d. This characteristic makes the BMICl‐electrode easier to store and more suitable for commercialization.

2.2. The Suppressed Shuttle Effects and I⁺ Hydrolysis

In addition to the BMICl in the cathode's effects on suppressing the dissolution and sublimation of iodine during rest periods, BMICl also mitigate the shuttle effects of polyiodides and the hydrolysis of I⁺ during cycling through the reactions shown in Figure 2a. The effectiveness in suppressing the shuttle effects was assessed using various techniques. The −NH⁺ moieties within the BMICl interact with polyiodides (I_x ⁻, where x = 3 or 5) to form insoluble BMICl‐CNTs‐I_x composites, as evidenced by the color change observed in solution following the addition of the BMICl to I₃ ⁻ solution (Figure S13, Supporting Information). Raman spectroscopy (Figure S14, Supporting Information) provided substantial evidence for this interaction, revealing that the solution containing I₃ ⁻ exhibits a prominent Raman band at 112 cm⁻¹ and a shoulder peak at 148 cm⁻¹, corresponding to the symmetric and anti‐symmetric vibrations of I₃ ⁻.^{[ 21 ]} The BMICl‐CNTs‐I_x composite, however, displayed distinct Raman bands for I₃ ⁻ (≈108 cm⁻¹) and I₅ ⁻(≈166 cm⁻¹),^{[ 20 ]} confirming the effective precipitation of various polyiodide species by BMICl‐CNTs. Additionally, ATR‐FTIR spectroscopy further confirmed the interaction between polyiodides and BMICl‐CNTs, with the peak of BMICl shifting from 1169 to 1164 cm⁻¹ following interaction with the I₃ ⁻ solution (Figure S15, Supporting Information).^{[ 32 ]}

Figure 2.

The interaction between iodine and BMICl. a) Reactions of BMICl with I₃ ⁻ and ICl₂ ⁻. b) Operando synchrotron piezo‐controlled ATR‐FTIR spectra of Zn–I₂ batteries with BMICl‐electrode. Operando UV–vis spectra of c) NaCl‐electrode and d) BMICl‐electrode in 2 m ZnSO₄+0.5 m NaCl electrolyte during the first discharge process. Operando UV−vis spectra of e) NaCl‐electrode and f) BMICl‐electrode in 2 m ZnSO₄+0.5 m NaCl electrolyte during the first charge process.

The interaction between BMICl and polyiodides during the cycling process was further investigated. Operando synchrotron piezo‐controlled ATR‐FTIR spectra were employed to monitor changes in the Zn–I₂ batteries with BMICl electrodes during the first discharge process (Discharge first, from I⁰ to I⁻). As depicted in Figure 2b, as the discharge progresses, peaks at ≈2876 cm⁻¹ (C‐H symmetric stretch mode) and 2915 cm⁻¹ (C‐H anti‐symmetric stretch mode) gradually emerge,^{[ 32 ]} indicating that the polyiodides is interacting with BMICl. Additionally, operando UV–vis spectroscopy was utilized to assess the dissolution of polyiodides during battery cycling. In the electrolyte of 2 m ZnSO₄+0.5 m NaCl, the absorbance of I₃ ⁻ at 285 and 350 nm increases steadily during the discharging, signifying substantial I₃ ⁻ dissolution (Figure 2c). This outcome aligns with the noticeable color change observed in the quartz battery after cycling (Figure S16a,b, Supporting Information). Remarkably, the incorporation of BMICl‐CNTs into the electrode significantly reduces the I₃ ⁻ dissolution, evidenced by much weaker I₃ ⁻ absorbance in Figure 2d and minimal color changes in the solution, as presented in Figure S16c,d (Supporting Information).

The impact of the BMICl‐CNTs composite within the electrode on the second‐step I⁰/I⁺ conversion was subsequently investigated. During the conversion reaction between I⁰ and I⁺, I⁺ tends to form ICl₂ ⁻ when Cl⁻ is introduced into the electrolyte. Similarly, BMICl reacts with ICl₂ ⁻ due to the −NH⁺ moieties, changing the color of the ICl₂ ⁻ solution from orange to light yellow, as observed in Figure S17 (Supporting Information). The ICl₂ ⁻ solution exhibits a sharp Raman band at ≈266.4 cm⁻¹,^{[ 33 ]} attributed to the I−Cl vibration of ICl₂ ⁻ (Figure S18, Supporting Information). Upon the addition of the BMICl‐CNTs composite, the intensity of this peak significantly decreases, indicating that BMICl has bonded with ICl₂ ⁻. This observation aligns with the changes in the ATR‐FTIR spectra, where the BMICl peak shifts from 1169 to 1163 cm⁻¹ (Figure S19, Supporting Information). In the hydrolysis process of I⁺, ICl₂ ⁻ initially reacts with H₂O, producing HIO and HCl.^{[ 34 ]} Subsequently, HIO can decompose into HI and HIO₃. During the charging process (Charge first, from I⁰ to I⁺), the HI is charged back to I₃ ⁻. Therefore, the concentration of I₃ ⁻ in the I⁰ to I⁺ process can also indicate the extent of I⁺ hydrolysis. UV–vis spectroscopy during the I⁰ to I⁺ process, as recorded in Figure 2e,f and Figure S20 (Supporting Information), shows a much higher concentration of I₃ ⁻ and deeper color in the NaCl‐electrode system, indicating that the BMICl‐electrode effectively suppresses I⁺ hydrolysis.

2.3. The Enhanced Reaction Kinetics of I⁰/I⁺ Redox Couple

The electrochemical performance of the 4eZIBs was evaluated initially in the coin‐cell configuration. As shown in the cyclic voltammetry (CV) profiles, the NaCl‐electrode‐based 4eZIBs exhibit a pronounced current response corresponding to the I⁻/I⁰ redox couple, while the I⁰/I⁺ redox couple is significantly weaker (Figure 3a, bottom). This observation suggests that the conversion of I⁰ to I⁺ is unfavourable even with NaCl in the electrode. This is caused by the high solubility of NaCl in the electrolyte, resulting in the loss of NaCl at the electrode. In contrast, the BMICl‐electrode‐based 4eZIBs display distinct current responses for I⁰/I⁺ redox couples (Figure 3a, top). The current response corresponding to the I⁻/I⁰ redox couple and the I⁰/I⁺ redox couple are nearly equal, confirming that the I⁰/I⁺ conversion is fully activated. This improvement is primarily attributed to the incorporation of a spatially confined Cl⁻ source within the electrode architecture, which facilitates localized Cl⁻ availability and enhances the overall reaction kinetics. To confirm the good transformation of I⁻/I⁰ redox couple and the I⁰/I⁺ redox couple, the operando Raman spectra recorded during battery operation were collected. Figure 3b presents the corresponding galvanostatic charge‐discharge (GCD) curves of 4eZIBs, while Figure 3c shows the operando Raman results upon battery cycling. During the first‐step I⁻/I⁰ conversion process, iodine undergoes stepwise transformation, forming I₃ ⁻ and I₅ ⁻ intermediates. When the potential exceeds 1.6 V, the Raman band for I₃ ⁻ gradually diminishes, and the Raman band for I₅ ⁻ shifts to a higher Raman shift, indicating the progressive conversion of polyiodides to I₂. As the voltage further increases beyond 1.8 V, the Raman band for I₂ disappears during the second‐step conversion, marking the complete oxidation of I₂.

Figure 3.

Electrochemical performance of 4eZIBs. a) CV profiles comparing 4eZIBs equipped with NaCl‐electrodes or BMICl‐electrodes. b) GCD profiles for operando Raman testing of BMICl‐electrode/Zn full batteries. c) Operando Raman contour map illustrating the spectral changes in BMICl‐electrode/Zn full batteries during operation. d) GITT plot recorded after the first discharge cycle. e) Diffusion coefficient versus voltage, computed from GITT data during the charge phase. f) Diffusion coefficient versus voltage, computed from GITT data during the discharge phase. g) GCD profiles of NaCl‐based 4eZIBs at 0.5C. h) GCD profiles of BMICl‐based 4eZIBs at 0.5C.

The effect of BMICl on the conversion kinetics of the I⁻/I⁰ and I⁰/I⁺ redox couples was further examined using the galvanostatic intermittent titration technique (GITT). The GITT plots were obtained at a current rate of 1C (Figure 3d; Tables S2 and S3, Supporting Information). During the charging process (Figure 3e), the diffusion coefficient (D) for the NaCl‐electrode system shows a significant decrease in the voltage range above 1.6 V, indicating a much slower conversion rate for I⁰/I⁺ compared to I⁻/I⁰. In contrast, the decrease in D for the BMICl‐electrode in the same voltage range is considerably smaller, demonstrating that BMICl enhances the conversion kinetics of I⁰/I⁺. Additionally, the abnormal increase in D between 1.9 and 1.6 V with the NaCl‐electrode is caused by the strong I⁺ hydrolysis in the discharge process (Figure 3f). The improved conversion kinetics in the BMICl‐electrode contribute to a higher discharge capacity. As shown in Figure 3g,h, the I⁰/I⁺ redox couple delivers a discharge capacity of ≈208 mA h g⁻¹ in BMICl‐electrode‐based 4eZIBs, significantly higher than the ≈125 mA h g⁻¹ observed in NaCl‐electrode‐based 4eZIBs.

2.4. The Electrochemical Performance of 4eZIBs

The rate performance of 4eZIBs in coin‐cell configuration was further evaluated (Figure 4a; Figure S21, Supporting Information). Analyzing the discharge capacity contribution from the I⁰/I⁺ redox pair, the NaCl‐electrode system delivers only 120, 100, 82, and 10 mA h g⁻¹ at 1C, 2C, 5C, and 10C, respectively. In contrast, even at a high rate of 10C, the BMICl‐electrode system maintains a substantial discharge capacity of 66.7 mA h g⁻¹, much higher than that of NaCl‐electrode based system (Figure 4b,c), indicating the much better rate performance of the BMICl‐electrode system. Regarding long‐term cycling stability, the BMICl‐electrode system demonstrates excellent durability, maintaining 93.4% of its initial capacity after 600 cycles at 1C with a high areal capacity over 3 mA h cm⁻² (Figure 4d; Figure S22, Supporting Information). In contrast, the NaCl‐electrode system experiences a sudden drop in CE after only 130 cycles due to the serious I⁺ hydrolysis. Moreover, the NaCl‐based system exhibited an average CE of 93.79%, while the BMICl‐based system achieved a significantly higher value of 99.37%, indicating the effective suppression of parasitic reactions such as polyiodide shuttle effects and I⁺ hydrolysis by the quasi‐solid additive. Furthermore, the BMICl‐electrode system sustains over 2 000 cycles at a high current density of 10C, whereas the control group suffers from a short circuit after ≈100 cycles due to the shuttle‐induced corrosion of the zinc anode (Figure 4e).^{[ 35 ]}

Figure 4.

Cycling performance of 4eZIBs. a) Comparison of the rate performance of 4eZIBs with different electrode configurations. b) Discharge capacity analysis of I⁻/I⁰ redox pair. c) Discharge capacity analysis of I⁰/I⁺ redox pair. d) Cycling stability of 4eZIBs at a low current rate of 1C. e) Long‐term cycling stability of BMICl‐electrode‐based 4eZIBs at 10C.

The BMICl‐electrode system was further scaled up from coin cells to practical pouch cells to assess the real‐world feasibility and robustness of the proposed electrode design strategy. Impressively, at a high current rate of 2C, the pouch cell delivers remarkable long‐term durability, maintaining 82% of its initial capacity after 500 cycles (Figure 5a). Even after such prolonged cycling, the I⁺ redox process continues to contribute ≈50% of the total capacity (Figure 5b), indicating that the BMICl remains highly effective and stable over extended operation. Even after increase the areal capacity, the pouch cell still maintained a high I⁺ conversion rate. As depicted in Figure 5c, the second‐step I⁰/I⁺ conversion in the BMICl‐electrode‐based pouch cell contributes a substantial discharge capacity of ≈420 mA h, which is comparable to that of the first‐step I⁻/I⁰ redox pair (470 mA h), highlighting the efficient utilization of iodine species in both redox steps. The nearly equivalent capacity contributions from these two redox couples not only validate the full four‐electron transfer mechanism but also indicate the minimized shuttle effect and enhanced reversibility enabled by the BMICl‐modified electrode. Importantly, the assembled ≈0.9 Ah pouch cell (Figure 5d; Figure S23, Supporting Information), can achieve a high areal capacity of 7.1 mA h cm⁻² for each side of the cathode, much higher than that of the previously reported value (Figure 5e).^{[ 18} , ²⁶ , ³⁶ , ³⁷ , ³⁸ , ^{39 ]} The pouch cell retains 95.8% of its initial capacity after 150 charge–discharge cycles, underscoring the structural integrity of the electrode and the chemical stability of the redox‐active species under super‐high mass loading conditions. To further evaluate the effectiveness of the quasi‐solid additive in suppressing shuttle effects and I⁺ hydrolysis in a practical setup, self‐discharge tests were performed on pouch cells (Figure 5f,g). After 48 h of resting at open‐circuit, the NaCl‐based system retained only 61.15% of its initial capacity, whereas the BMICl‐based system maintained 86.04%, confirming the additive's role in mitigating polyiodide shuttle effects and I⁺ hydrolysis.

Figure 5.

Pouch cell performance of BMICl‐electrode‐based batteries. a) Cycling performance of the pouch cell at ≈2C (inset is the digital photo of the pouch cell). b) GCD profiles of the pouch cell at the 1^st cycle and 500^th cycle. c) GCD profiles of the pouch cell cycled at ≈0.5C. d) Cycling performance of the pouch cell at ≈0.5C (inset is the digital photo of the pouch cell). e) The areal capacity comparison of this work with previously reported work. Self‐discharge test of f) NaCl‐based pouch cell and g) BMICl‐based pouch cell after 48 h of resting. [Correction added on November 11, 2025, after first online publication: Figure 5 has been updated.]

Beyond electrochemical metrics, the pouch cell exhibits excellent mechanical durability and safety performance. As shown in Figure S24 (Supporting Information), the cell with the BMICl electrode is capable of steadily powering a red light‐emitting diode (LED), confirming its energy output stability. Even when physically damaged, for instance, upon being deliberately cut, the voltage drop remains as low as 0.051 V (Figure S25, Supporting Information). Remarkably, the device continues to operate and power a digital meter, as demonstrated in Figure S26 (Supporting Information). This exceptional tolerance to mechanical abuse reflects the intrinsic safety advantages of aqueous electrolyte systems and the robustness of the electrode stability enabled by the quasi‐solid BMICl‐CNTs. Together, these results strongly support the potential of BMICl‐electrode‐based Zn–I₂ pouch cells for practical energy storage applications, especially in scenarios demanding both high energy density and operational safety.

3. Conclusion

A quasi‐solid electrode additive strategy was developed to simultaneously address the major challenges associated with both the I⁻ to I⁰ and I⁰ to I⁺ redox conversions in 4eZIBs. By co‐grinding BMICl with CNTs, a quasi‐solid electrode additive is formed through strong π–π interactions. This additive effectively immobilizes polyiodide species, suppressing shuttle effects in the redox reaction of I⁻ to I⁰, while providing a stable and localized source of chloride ions within the electrode to accelerate I⁺ conversion kinetics and inhibit I⁺ hydrolysis in the second step of I⁰ to I⁺ redox conversion. As a result, the modified system enables efficient utilization of the full four‐electron iodine chemistry. The coin cells achieve a high specific capacity of 418.9 mA h g⁻¹ at 0.5C, approaching the theoretical value. The cells with the electrode additive can maintain 93.4% of their initial capacity after 600 cycles at 1C with a high cathode areal loading of over 3 mA h cm⁻². Moreover, when scaled up to pouch cells with a super‐high cathode areal capacity of 7.1 mA h cm⁻², the system still delivers stable performance, retaining 95.8% of its capacity after 150 cycles at a current density of 6.3 mA cm⁻². The pouch cells also exhibit excellent mechanical stability and safety, maintaining power output even after mechanical damage. This study demonstrates an effective method to enhance the performance of 4eZIBs, offering a promising approach to expanding the design horizon of advanced batteries for practical applications in aqueous battery technologies.

4. Experimental Section

Chemical Reagents

Zn foil, three‐electrode cell, Cu foil, aluminium‐plastic film, Ag/AgCl electrode, and hydrophobic polytetrafluoroethylene (PTFE) were purchased from Shenzhen Kejing Star Technology. ZnSO₄·7H₂O (≥99.0%), iodine (≥99.8%), sodium chloride (≥99%), and 1‐butyl‐3‐methylimidazolium chloride (≥98.0%) were purchased from Sigma‐Aldrich Chemical Co. Ketjen Black (KB) and carbon nanotubes were purchased from Guangdong Canrd New Energy Technology Co. All other reagents were analytical grade and used directly without purification. Deionized water was used to prepare all aqueous electrolytes.

Electrode Preparation

First, I₂ (50 wt.%) and KB (50 wt.%) were ground in a mortar for 30 min and then vacuum‐dried at 80 °C for 3 h to obtain I₂@KB. Separately, 0.13 g of CNTs was mixed with varying amounts of BMICl—0.14 g (corresponding to an I:Cl molar ratio of ≈4:1), 0.28 g (corresponding to an I:Cl molar ratio of ≈2:1), or 0.56 g (corresponding to an I:Cl molar ratio of ≈1:1)—and ground for 30 min to form a gel‐like composite. Subsequently, 0.8 g of I₂@KB was added to the mixture and ground for another 30 min to ensure uniform distribution. Finally, 0.5 g of PTFE solution (10 wt.%) was incorporated and homogenized to form the electrode slurry. This slurry was then cast onto a Ti mesh, setting the stage for subsequent electrode characterization and testing (The fabrication process for the NaCl‐electrode follows the same procedure as that for the BMICl‐electrode, with the substitution of 0.093 g NaCl in an equivalent molar ratio to BMICl). For coin cell, the areal loading of active materials was ≈8 mg cm⁻². For the pouch cell, the areal loading of active materials was ≈17.5 mg cm⁻².

Electrolyte Preparation

A 2 m ZnSO₄ electrolyte was prepared using ZnSO₄·7H₂O and deionized water in a 100‐mL volumetric flask. Based on aqueous ZnSO₄ electrolyte, 0.5 m of NaCl was added to establish electrolytes of 2 m ZnSO₄ + 0.5 m NaCl.

Assembly of Pouch Cells

The assembly process for pouch cells involved several steps. First, cathode electrodes were prepared by above mentioned method and pressed onto two sides of Ti mesh (70 × 90 mm), followed by drying at 40 °C. Subsequently, ≈0.2 A h pouch cell was assembled by stacking one cathode with one Zn foil (100 µm, 72 × 92 mm). Glass fiber (Whatman, GF/D) and a 2 m ZnSO₄+0.5 m NaCl aqueous solution served as the separator and electrolyte, respectively. The thickness of the assembled pouch cell was ≈0.3 cm. The added electrolyte volume was controlled at ≈10 mL Ah⁻¹. Subsequently, ≈0.9 A h pouch cell was assembled by stacking one cathode (two sides coated) with two Zn foils (100 µm, 72 × 92 mm). Glass fiber (Whatman, GF/D) and a 2 m ZnSO₄+0.5 m NaCl aqueous solution served as the separator and electrolyte, respectively. The thickness of the assembled pouch cell was ≈0.4 cm. The added electrolyte volume was controlled at ≈10 mL Ah⁻¹ as well.

Characterizations

Cyclic voltammetry curves (CV) were determined using a VMP3 (CHI 760E, Chenhua). The galvanostatic cycling studies were determined using the LAND battery testing system at room temperature (25 °C). The working voltage of the cells was set from 0.6 to 1.9 V versus Zn/Zn²⁺ (for the high‐rate performance, the voltage range has been set to 0.6–2.0 V). Attenuated total reflection Fourier‐transform infrared (ATR‐FTIR) spectroscopy was performed with a Thermo‐Fisher Nicolet iS20 equipped with a liquid, nitrogen‐cooled HgCdTe (MCT) detector using a VeeMax III ATR accessory (Pike Technologies). The battery was tested by NEWARE battery test system (CT‐3008‐5V1mA‐164, Shenzhen, China). A UV–vis spectrophotometer (UV2600, Shimadzu, Japan) was used to obtain the operando UV–vis diffuse reflectance and UV–vis absorption spectra. Operando Raman spectroscopy for the Zn batteries was determined using a confocal Raman microscope (Horiba LabRAM HR Evolution) with a 50X (1.0 N.A) objective (Olympus). The laser wavelength was 532  nm.

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

S.‐Z.Q. and J.H. supervised the project. H.W. designed experiments and conducted characterizations and electrochemical measurements. J.V. developed the piezo‐controlled ATR‐FTIR technique and S.‐J.Z. and Y.J. conducted partial characterizations of materials. H.W. and J.H. analysed data and wrote the manuscript. S.‐Z.Q. reviewed and revised the manuscript.

Supporting information

Supporting Information

Wu H., Zhang S.‐J., Vongsvivut J., Jiang Y., Hao J., and Qiao S.‐Z., “Quasi‐Solid Cathode Additive Enables Highly Reversible Four‐Electron I⁻/I⁰/I⁺ Conversion in Aqueous Zn‐I₂ Batteries.” Adv. Mater. 38, no. 3 (2026): e11680. 10.1002/adma.202511680

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.