Halogen-confining host materials for high-performance zinc–halogen batteries

Abstract

Zinc–halogen batteries hold great promise for grid-scale energy storage owing to their multi-electron transfer capability, abundant halogen resources, low cost and high theoretical voltage and capacity. However, they are still constrained by the sluggish redox kinetics of halogen species and the uncontrollable shuttling of polyhalide intermediates, which compromise energy efficiency and cycling stability. In this regard, the rational design of halogen-confining host materials has emerged as a promising strategy; however, a comprehensive review of this fast-evolving field is still lacking. This tutorial review begins with an overview of configurations and fundamental mechanisms of zinc–halogen batteries, followed by in-depth discussions on their thermodynamic and kinetic characteristics governing halogen reactions. We then critically analyze the key challenges of halogen cathodes and propose a confinement–catalysis–conduction triad to rationalize the design of host materials, elucidating their structure–performance correlations and mechanistic insights across various zinc–halogen battery systems, including Zn–Cl₂, Zn–Br₂, Zn–I₂ and Zn–dual halogen configurations. Furthermore, optimization strategies encompassing rational structural design, surface functionalization, heteroatom doping, engineering of single/dual-atom catalysts and heterostructure engineering are highlighted to promote halogen confinement, accelerate redox kinetics, and facilitate charge transport within halogen-based cathodes. Finally, we provide a concise perspective on existing barriers and emerging opportunities, offering valuable guidance for high-performance zinc–halogen batteries.

---

Shude Liu

Shude Liu received his PhD degree from the School of Mechanical Engineering, Yonsei University, South Korea (2020). He was a JSPS postdoctoral researcher at the National Institute for Materials Science, Japan (2021–2023). After that, as a professor, he joined Donghua University (2023). His research interests primarily focus on the design and development of key materials for flexible electrochemical energy storage and conversion.

Bin Ding

Bin Ding is currently a Professor at Donghua University, China. He received his BS from Northeast Normal University, China, in 1998 and his MS from Chonbuk National University, South Korea, in 2003. He subsequently earned his PhD degree from Keio University, Japan, in 2005, followed by a postdoctoral fellowship at the University of California, Davis, USA. Prof. Ding formerly served as the Dean of the Institute of Science and Technology at Donghua University. Since 2025, he has served as the Vice President of Shanghai Polytechnic University, China. His research mainly focuses on the controllable fabrication, functionalization, and industrial applications of fibrous materials, which have found extensive use in thermal insulation, warmth retention, filtration, waterproof and breathable membranes, and energy electronics.

Yusuke Yamauchi

Yusuke Yamauchi received his PhD degree from Waseda University, Japan, in 2007. Following that, he joined the National Institute for Materials Science, Japan. Since 2016, he has been a full professor at the School of Chemical Engineering and a senior group leader at AIBN at The University of Queensland. Since 2023, he has also served as a distinguished full professor at Nagoya University. Additionally, he is an associate editor of J. Mater. Chem. A (RSC) and Chem. Eng. J. (Elsevier). He has authored over 1000 papers with over 100 000 citations (h-index > 170). He has been recognized as a Highly Cited Researcher in Chemistry and Materials Science and the JST-ERATO Research Director (2020–), as well as an Australian Laureate Fellow (2023).

Seong Chan Jun

Seong Chan Jun is currently a Professor at Yonsei University, South Korea. He received his BS, MS, and PhD degrees from George Washington University, Cornell University, and Columbia University, respectively. In 2022, he was elected as a Fellow of the Korean Academy of Science and Technology. His current research is focused on the design of novel nanomaterials and their applications in electronics, photonics, and flexible energy conversion and storage technologies.

---

1. Introduction

The global energy crisis and environmental pollution issues have become increasingly severe, necessitating the development of clean energy sources such as solar, wind, and tidal energy sources.¹ However, the intermittent and fluctuating nature of renewable energies hinders their ability to provide a stable power supply, which underscores the need for efficient and reliable energy storage technologies.² Lithium-ion batteries (LIBs) have dominated the market for decades because of their high energy density and long cycle life, making them the primary choice for electric vehicles, portable electronic devices, and smart grids.³ However, LIBs rely on organic solvent-based electrolytes, which are flammable and prone to thermal runaway under conditions of overcharging, deep discharging, or thermal shock, posing significant safety risks.⁴ In addition, the scarcity and uneven geographical distribution of lithium resources hinder their widespread deployment for energy storage.⁵ Therefore, there is a pressing demand for developing safe, durable, and cost-effective alternative battery systems that can satisfy the growing market requirements.⁶ In contrast, aqueous batteries utilizing neutral solutions as electrolytes offer significant advantages, including high safety, environmental friendliness, and high ionic conductivity.^7–9 Among them, zinc-based batteries have garnered great attention because of the abundant availability of zinc, low cost, high theoretical capacity (5855 mAh cm⁻³ and 820 mAh g⁻¹), low oxidation potential (−0.76 V vs. the standard hydrogen electrode (SHE)), and high compatibility with aqueous electrolytes, which is highly competitive for other metal-based batteries (Fig. 1a).^10–12 These attributes make aqueous zinc-based batteries a promising and viable alternative compared to commercial LIBs.

Fig. 1 (a) Comparative analysis of performance parameters among Zn and other metal-based batteries, including molar mass, volumetric capacity, standard potential, gravimetric capacity, density and abundance in the crust. (b) Comparison of Zn–Cl₂, Zn–Br₂ and Zn–I₂ batteries in terms of their electrochemical properties (e.g., self-discharge rate, theoretical voltage, cycling life and redox potential) and (c) commercialization potential (e.g., industry maturity, cost of halogen raw materials, system complexity and security risks). (d) Publication trends of zinc–halogen batteries over the last decade. Data were collected from the Web of Science using the keywords “zinc–halogen (Cl₂, Br₂ and I₂) batteries” collected up to June 2025 (the inset is the proportion of publications on Zn–Cl₂, Zn–Br₂, and Zn–I₂ batteries from 2015 to 2025).

In the past few decades, significant progress has been made in the development of cathode materials (e.g., Prussian blue analogues (PBAs), vanadium-based compounds, manganese-based compounds and organic compounds) for aqueous zinc-based batteries,^13–16 particularly in enhancing their energy storage capacity and cycling stability. However, most of these materials rely on surface redox mechanisms or ion intercalation reactions, often suffering sluggish Zn²⁺ diffusion, structural degradation, and insufficient energy density.^17,18 Moreover, these cathodes operate through a Zn²⁺/H⁺ co-intercalation mechanism;⁹ however, the insertion of H⁺ leads to interfacial pH variations and by-product accumulation, which inevitably compromise their structural stability and electrochemical performance.¹⁹ To this end, recent advancements have focused on exploring alternative materials with improved electrochemical properties. Recently, halogen-based cathode materials (Cl₂, Br₂, and I₂) have gained increasing research attention because of their abundant material availability, reversible halogen conversion reactions, wide potential windows (Cl⁻/Cl⁰: 1.35 V; Br⁻/Br⁰: 1.08 V; I⁻/I⁰: 0.53 V (vs. SHE)) and high theoretical specific capacities (Cl⁻/Cl⁰: 756 mAh g⁻¹; Br⁻/Br⁰: 335 mAh g⁻¹; I⁻/I⁰: 211 mAh g⁻¹; I⁻/I⁺: 422 mAh g⁻¹; I⁻/IO₃⁻: 1266 mAh g⁻¹),^20–22 which make them promising candidates for next-generation energy storage technologies.²³ The development of zinc–halogen batteries can be traced back to as early as 1884, when Charles Renard first pioneered the concept of a Zn–Cl₂ liquid battery.²⁴ Although the prototype exhibited limited energy density and rechargeability, it sparked ongoing exploration and innovation in the development of novel battery chemistries. In the following decades, Zn–Br₂ and Zn–I₂ batteries were successively developed with a focus on enhancing capacity and cycling stability, leading to notable progress and renewed interest in zinc–halogen systems for large-scale energy storage.^25,26 Additionally, we present a comparative analysis of different types of zinc–halogen batteries in terms of their electrochemical properties (e.g., self-discharge rate, theoretical voltage, redox potential and cycling life) and commercialization potential (e.g., cost of halogen raw materials, system complexity, security risks and industry maturity), as shown in Fig. 1b and c. This comparison highlights the distinct advantages and trade-offs associated with each zinc–halogen technology. Nevertheless, zinc–halogen batteries still face the shuttle effect, caused by the formation of polyhalides during charge–discharge cycles and the high solubility of halogen species in the electrolyte,^27,28 leading to zinc corrosion and irreversible structural degradation. Furthermore, the low electrical conductivity of halogen cathodes induces their accumulation on the electrode surface,²⁹ forming a passivation layer that severely impedes redox kinetics, thereby reducing the efficiency of halogen utilization and degrading the electrochemical performance of zinc–halogen batteries.

Recently, confining halogens within suitable host materials has been demonstrated to be effective, primarily due to their strong adsorption ability and good charge conduction for halogen species,^30,31 which reduce halogen solubility in the electrolyte, suppress the shuttle effect, and enhance electrochemical catalytic kinetics. Thus far, researchers have developed various halogen-confining host materials including carbonaceous materials,³² functional frameworks and their derivatives,³³ two-dimensional (2D) transition metal carbides/nitrides, non-framework polymer structures, and transition metal sulfides.³⁴ Specifically, carbonaceous materials have significant potential for halogen confinement due to their high electrical conductivity and extensive specific surface area.³¹ Nanoporous carbon architectures are particularly effective in suppressing polyhalide diffusion and mitigating the shuttle effect. However, their dependence on physical adsorption limits polyhalide stabilization efficacy, while prolonged cycling may cause material degradation via oxidative corrosion or structural collapse. Functionalized framework structures featuring large surface areas and abundant functional groups offer superior halogen adsorption and catalytic activity, yet they remain vulnerable to degradation under highly humid or strongly acidic conditions, compromising their cycling stability.³⁵ The 2D transition metal carbides/nitrides exhibit unique advantages for halogen stabilization through their layered structures and exceptional conductivity. However, extended cycling may induce unfavorable interfacial reactions with aqueous electrolytes, gradually deteriorating performance.³⁶ Non-framework polymeric materials, such as starch, serve as cost-effective, biodegradable alternatives for scalable applications, but their inherently poor conductivity hinders high-performance implementations. Transition metal sulfides generally provide good electrical conductivity but are prone to significant structural instability during halogen intercalation. Additionally, irreversible surface reactions with polyhalides further compromise material utilization and reduce battery efficiency. These limitations illustrate the necessity for continued progress in developing optimal host materials for zinc–halogen batteries. Fig. 1d illustrates the rapid growth of research on zinc–halogen batteries, particularly in host material engineering, reflecting rising interest. This underscores the need for a timely and comprehensive review that systematically categorizes halogen-confinement strategies, identifies major challenges, and offers design insights for next-generation host materials to enable high-performance zinc–halogen batteries.

To date, several reviews have summarized the research progress of zinc–halogen battery systems. For example, Zhi et al.²² reviewed the progress of static conversion batteries from a microscopic chemical perspective, focusing on halogen electrode reactions. While their discussion on the interplay between halogen reactions and other battery components is insightful, it lacks in-depth analysis of halogen-confining host materials and is limited to static battery configurations. This narrow scope makes it difficult to fully assess the development and challenges of cathode host materials in other configurations, such as flow batteries. Similarly, Lu et al.³⁷ summarized the electrochemical mechanisms of zinc–halogen batteries and addressed the progress of individual components (cathodes, anodes, membranes and electrolytes), but did not specifically discuss halogen host materials, nor did they provide detailed optimization strategies or mechanistic analyses of performance improvement. Neither of these two reviews offered a systematic categorization of cathode host materials across different zinc–halogen systems. Wang et al.²⁷ briefly mentioned various conversion-type electrode materials, including halogens, MnO₂, chalcogenides, and Cu-based compounds, in the context of zinc aqueous metal batteries in non-alkaline electrolytes. However, their discussion of cathode materials was limited, with halogen-confining materials only briefly mentioned and Zn–Cl₂ systems excluded, leading to an incomplete understanding of host material development in zinc–halogen batteries. Hu et al.³⁸ focused on Zn–I₂ systems, discussing iodine redox mechanisms and the main challenges of cathodes, anodes, electrolytes, and membranes. However, their treatment of host materials lacked classification by optimization strategies and did not address Zn–Cl₂ or Zn–Br₂ systems, limiting the generalizability of their conclusions. Similarly, Ma et al.³⁹ provided an in-depth and systematic review of Zn–I₂ batteries, highlighting both low- and high-valent iodine redox mechanisms, electrolyte engineering, and device-level strategies. Their work emphasized the multielectron conversion of high-valent species (e.g., I⁺/IO₃⁻), offering valuable mechanistic insights. However, the review does not include Zn–Cl₂ or Zn–Br₂ systems, nor does it present a structured discussion on halogen-confining host materials or their optimization strategies. Furthermore, these reviews either focus on a single halogen system or are limited in material-level discussions, making it difficult to gain a comprehensive understanding of cathode host design across the full range of zinc–halogen battery chemistries. Recently, Yang et al.⁴⁰ summarized the development of halogen host materials and discussed basic redox mechanisms and challenges. However, they did not systematically classify the host materials by types or optimization strategies, nor did they incorporate thermodynamic or kinetic considerations. Notably, they did not address Zn–dual halogen systems, which are gaining increasing attention but remain underexplored. Furthermore, Fang et al.⁴¹ focused on thermodynamic and kinetic regulation mechanisms in aqueous conversion-type batteries, providing a theoretical framework to understand key factors such as Gibbs free energy, ion diffusion barriers, and charge-transfer resistance in Zn–X (X = I₂, S, Br₂) systems. However, the review provided limited discussion on the structural design, functionalization, and performance-driven optimization of halogen-confining host materials and lacked a systematic classification or structure–property analysis essential for guiding practical material design. Nevertheless, these reviews either concentrate on specific halogen chemistry, provide general discussions on battery components, or emphasize theoretical modelling without addressing the critical role of halogen-confining host materials. Notably, recent advances have expanded host design paradigms beyond conventional structural and chemical modifications. These include single-/dual-atom catalysts (SACs/DACs), where isolated metal atoms undergo reversible coordination changes under operating conditions, enabling atom-efficient catalytic confinement with tunable activity; interdisciplinary operando strategies that integrate in situ X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and Raman spectroscopy with advanced data analytics, allowing real-time elucidation of electrochemical mechanisms to guide the rational design of host architectures; and artificial intelligence (AI)-assisted materials discovery, which utilizes high-throughput density functional theory (DFT) databases and machine-learning surrogate models to rapidly screen vast compositional and structural spaces and identify optimal host chemistries. Therefore, a comprehensive review is essential to address this gap, covering the working mechanisms of zinc–halogen batteries, the classification and advantages of halogen-confining host materials, the confinement–catalysis–conduction triad, structure–performance relationships, existing challenges, and optimization strategies for various types of halogen-confining host materials, as well as emerging issues and promising research directions for practical applications.

Given this context, this review provides an overview of the significant advancements in halogen-confining host materials for zinc–halogen batteries in recent years (Fig. 2).^{28,31,33,34,42–53} We begin by categorizing the configurations and working mechanisms of zinc–halogen batteries, emphasizing the thermodynamic and kinetic behaviors of the conversion reactions in halogen cathodes. Subsequently, we discuss the key challenges of halogen-confining host materials, propose the confinement–catalysis–conduction design concept for host materials, and explore the structure–performance relationships of typical host materials, such as carbonaceous materials, functional frameworks and their derivatives, as well as 2D transition metal carbides/nitrides, across various zinc–halogen battery systems. Additionally, we recommended several optimization strategies for improving the electrochemical performance of halogen-based cathodes. Finally, we conclude with a brief outlook, providing key barriers and insights to guide the development of high-performance zinc–halogen batteries.

Fig. 2 Process of some representative halogen-confining host materials for zinc–halogen batteries, where orange, green and purple lines represent Zn–Cl₂, Zn–Br₂ and Zn–I₂ batteries, respectively. Arranged chronologically along the path from top to bottom. Reproduced with permission from ref. 28, 31, 33, 34 and 42–53. Copyright 2019, Wiley-VCH. Copyright 2019, Wiley-VCH. Copyright 2020, Wiley-VCH. Copyright 2020, American Chemical Society. Copyright 2021, Wiley-VCH. Copyright 2021, Wiley-VCH. Copyright 2022, American Chemical Society. Copyright 2021, Elsevier. Copyright 2023, Elsevier. Copyright 2023, Royal Society of Chemistry. Copyright 2023, Wiley-VCH. Copyright 2023, Wiley-VCH. Copyright 2024, Royal Society of Chemistry. Copyright 2023, Wiley-VCH. Copyright 2024, Wiley-VCH. Copyright 2024, Wiley-VCH.

2. Configuration, fundamental principles and thermodynamic and kinetic behaviors of zinc–halogen batteries

2.1. Configuration of zinc–halogen batteries

Zinc–halogen batteries are classified into zinc–halogen flow and zinc–halogen static batteries based on their structure, as indicated in Fig. 3a and b.^27,41 Both types consist of a zinc anode and a halogen cathode with a halogen-confining host material, an electrolyte, and a separator. In static batteries, the electrolyte is stored within the battery, with halogen species loaded onto the host materials, eliminating the need for flowing electrolytes.⁵⁴ In contrast, flow batteries store the electrolyte in two external tanks. During operation, two circulation pumps drive the electrolyte into the cell. Halogen and zinc ions circulate between the electrolyte tanks and cell, undergoing redox reactions at the electrodes to generate voltage. Both batteries have their advantages and suitable application scenarios. Zinc–halogen flow batteries offer several benefits, including uniform mixing, the active materials are thoroughly mixed through circulating flow or stirring to promote uniformity and stability of the reaction throughout the system; long cycle life, wherein the electrochemical reactions of active materials in flow batteries exhibit low polarization and high reversibility with minimal self-discharge issues, thereby contributing to extended storage lifetimes;⁵¹ and scalable capacity, where the power density is determined by the number and size of stacked units, while the capacity density is affected by the volume concentration of the electrolyte. Therefore, increasing the size and number of stacks can enhance the output power and increasing the electrolyte volume concentration can improve the energy storage capacity. Batteries with this configuration contribute to the realization of large-scale energy storage systems, meeting higher power and capacity demands. Static batteries exhibit advantages of cost efficiency, which implies that there is no need for flowing electrolytes or ion-exchange-type membranes, and a simple configuration, which implies that static batteries feature a straightforward design that facilitates easy operation without the need for stirring or circulation systems. These characteristics make static batteries ideal for portable electronic devices and flexible batteries.

Fig. 3 Structural schematics of (a) zinc–halogen flow batteries and (b) zinc–halogen static batteries.

2.2. Fundamental principles of zinc–halogen batteries

The energy storage mechanism of anodes in zinc–halogen batteries relies on Zn²⁺/Zn conversion, which involves the reversible electrodeposition and stripping of zinc metal.⁵⁵ During charging, Zn²⁺ ions near the anode gain electrons, are reduced, and deposit onto the anode surface. During discharge, the deposited Zn is oxidized to Zn²⁺ by losing electrons, which then migrates into the electrolyte under the influence of the electric field, while the released electrons flow to the halogen cathode via the external circuit. On the other hand, halogen-based cathodes rely on the reversible redox conversion of halogen species. According to their reaction pathways, these processes can typically be classified into two types of electron transfer mechanisms: the conventional X⁻/X⁰ and X⁻/X⁰/X⁺ conversions.

2.2.1. X⁻/X⁰ electron transfer mechanism. The conventional reaction mechanism at the cathode of zinc–halogen batteries involves the electrochemical reaction of the typical X⁻/X⁰ redox couple.⁵⁶ In this process, the halogen cathode undergoes reversible conversion between halide ions (X⁻) and neutral halogen species (X⁰) through single-electron transfer, enabling energy storage and release. The overall cathodic reaction in zinc–halogen batteries can be summarized as:

Anode: Zn²⁺ + 2e⁻ ⇌ Zn

Cathode: 2X⁻ ⇌ 2X⁰ + 2e⁻

Overall: ZnX₂ ⇌ Zn + 2X⁰

However, during the charge and discharge process of zinc–halogen batteries, the conversion between X⁻ and X⁰ occurs in a stepwise manner. X⁰ tends to undergo a series of side reactions with X⁻ to form soluble intermediates X_n+2⁻ (2X⁰ + X_n⁻ ⇌ X_n+2⁻, n = 1, 3, 5, 7…).⁵⁷

2.2.2. X⁻/X⁰/X⁺ electron transfer mechanism. Halogens possess variable valence states and can be oxidized to higher oxidation states. Leveraging the multielectron transfer chemistry of halogens can fundamentally enhance electrochemical performance for achieving a substantial increase in capacity.⁵⁸ Constructing interhalogen bonds (I–Br, I–Cl, and Br–Cl) plays a pivotal role in enabling the X⁻/X⁰/X⁺ redox transitions by promoting interhalogen conversion. Taking the typical I⁻/I⁰/I⁺ redox reaction in Zn–I₂ batteries as an example, the oxidation and reduction of I⁻/I⁰/I⁺ can be triggered when the applied voltage exceeds 1.8 V vs. Zn²⁺/Zn. However, this multistep redox process requires the presence of nucleophilic species (e.g., F⁻, Cl⁻ and Br⁻), which interact with the electrophilic I⁺ to form halogen–halogen complexes that stabilize the I⁻/I⁰/I⁺ transitions. Among these, Cl⁻ exhibits relatively strong nucleophilicity, which can form stable yet redox-active species such as ICl. Consequently, Cl is introduced to activate the I⁻/I⁰/I⁺ redox reaction and stabilize iodine species, which enables the release of a high theoretical specific capacity of 422 mAh g⁻¹.⁵⁹ The electrochemical reactions involving the formation of I–Br and I–Cl interhalogen bonds for stabilizing iodine species are summarized as follows:⁶⁰

Anode: Zn²⁺ + 2e⁻ ⇌ Zn

Cathode: 2I⁻ ⇌ 2I⁰ + 2e⁻

2I⁰ + 2X⁻ ⇌ 2IX + 2e⁻

Overall: Zn²⁺ + 2X⁻ ⇌ Zn + 2X⁰

Zn²⁺ + I⁰ + 2X⁻ ⇌ Zn + 2IX

2.3. Thermodynamic and kinetic behaviors governing halogen reactions

The thermodynamic and kinetic characteristics fundamentally determine the energy output and efficiency of zinc–halogen batteries employing conversion-type reactions. Thermodynamic factors involve the working potential, capacity, stability, and reversibility of the conversion process (Fig. 4a).⁴¹ The stability and reversibility of the electrode affect the coulombic efficiency (CE) and energy efficiency of batteries. The thermodynamic challenges faced by the cathode side of the battery include oxygen evolution, dissolution of halogen active species, and phase change. The conversion reaction kinetics are affected by charge transfer resistance, active ion diffusion rate, and phase change energy barrier. Mitigating the charge transfer resistance and phase transition barrier, while improving ion diffusion, can significantly enhance kinetic performance (Fig. 4b). Halogen redox processes require reversible transitions between solid and liquid phases, leading to both solid- and liquid-phase reactions.⁶¹ For the solid-phase reactions, stability is affected by the reversibility of the conversion products, whereas the reversibility of the reaction is affected by the electrode conductivity, interface characteristics of the electrode/electrolyte, and passivation products.⁶² Furthermore, solid-phase reactions need to overcome the high energy barriers arising from the stable chemical bonds of solid reactants and the low electronic conductivity of most solid phases (Fig. 4c). For liquid-phase reactions, stability is influenced by factors such as the charge transfer and diffusion coefficients of the liquid reactants, the shuttling effect, the number of current collectors and active sites, and the consumption of active ions (Fig. 4d).

Fig. 4 Key factors influencing (a) thermodynamic behavior, (b) kinetic behavior, (c) solid-phase reactions, and (d) liquid-phase reactions in conversion-type aqueous zinc-based batteries.⁴¹ Reproduced with permission from ref. 41. Copyright 2024, American Chemical Society.

3. Halogen-confining host materials for zinc–halogen batteries

The performance bottleneck of zinc–halogen batteries primarily arises from challenges encountered by the halogen cathodes during charge–discharge processes. (i) Dissolution and shuttle effects of halide ions: halogen species X (such as Br or I) react with halide ions X⁻ in the aqueous electrolyte during charge and discharge cycles, forming highly soluble polyhalide ions (such as Br₃⁻, Br₅⁻, I₃⁻, and I₅⁻),⁶³ as shown in Fig. 5a. As the reaction progresses, these polyhalide ions gradually accumulate on the electrode surface. Eventually, under the influence of the electric field and concentration gradients, they shuttle across the separator to the anode. This shuttle effect leads to reactions with zinc, deteriorating the electrode/electrolyte interface, causing zinc anode corrosion and passivation.⁶⁴ Consequently, uneven zinc deposition occurs, leading to capacity fading, a significant decrease in CE, and severe self-discharge issues, which ultimately deteriorate battery performance. (ii) Polarization and sluggish redox kinetics: overcoming the activation energy barrier in zinc–halogen batteries is crucial for initiating halogen-based conversion reactions.⁴¹ This barrier manifests as polarization overpotentials during charging and discharging (Fig. 5b and c). Under high-rate charge/discharge conditions, the activation energy barrier increases significantly due to sluggish cathodic reaction kinetics, resulting in elevated overpotentials.⁶⁵ This phenomenon increases the charging voltage (energy input) and decreases the discharging voltage (energy output), reducing voltage efficiency and energy efficiency as the charge/discharge rate increases, ultimately limiting the power density. The sluggish redox kinetics of halogen-active species can be attributed to limited electron transport and hindered mass transfer.⁴⁰ In terms of electron transport, the inherently low electrical conductivity and electrochemical inertness of halogen species severely impede efficient electron transport.²² In terms of mass transfer, the majority of the pores in the host material are often blocked by halide ions, limiting the effective diffusion of halogen-active species.⁴⁰ These issues result in slow reaction kinetics, leading to high polarization, low power density, and limited cycle life.

Fig. 5 Schematic diagrams of the issues in zinc–halogen conversion-type batteries: (a) halogen species dissolution, shuttle effect and typical problems of zinc anodes, (b) high reaction barrier, and (c) electrode polarization.

To address these key challenges of halogen-based cathodes, including halogen dissolution and shuttle effects, sluggish redox kinetics and sluggish charge transport, we summarize and propose a confinement–catalysis–conduction concept for designing host materials (Fig. 6), based on recent developments in host materials for zinc–halogen batteries. This design paradigm systematically optimizes the functionality of host materials by employing spatial confinement to suppress the halogen shuttle, introducing catalytically active sites to accelerate redox reactions,⁶⁶ and constructing conductive networks to enhance electron/ion transport efficiency, thereby improving the electrochemical performance of zinc–halogen batteries. Specifically, first, the confinement effect is achieved by constructing structures with nanoscale channels, interlayer gaps, or spatial confinement sites, or by introducing specific chemical sites or functionalized surfaces in the host material,⁶⁷ which chemically adsorb or react with halogen species (such as I₃⁻, I₅⁻, and Br₃⁻). This effectively inhibits the dissolution and diffusion of halogen and polyhalide anions, mitigates the shuttle effect, and stabilizes the intermediate conversion path. Second, chemical catalysis is achieved by introducing single-atom sites with high coordination unsaturation, M–N₄ structures, or Lewis acidic centers into the host material,⁶⁸ which promotes electron redistribution and activation of the conversion path during halogen redox processes, significantly lowering the reaction Gibbs free energy change (ΔG) and activation energy²² and accelerating halogen ion conversion kinetics. Finally, charge conduction relies on constructing a highly continuous conductive network to ensure efficient charge transport within cathodes.⁶⁹ Additionally, designing an appropriate pore size distribution and introducing surface hydrophilic functional groups can enhance the diffusion flux of Zn²⁺ ions and improve ion affinity at the electrode/electrolyte interface,⁷⁰ thus achieving high-efficiency charge transport. This design concept offers forward-looking and practical guidance for engineering host materials in next-generation zinc–halogen batteries.

Fig. 6 Design principle of halogen-confining host materials based on the confinement–catalysis–conduction paradigm.

In the analysis of practical zinc–halogen battery systems, to investigate the relationship between the porosity (ε) of the host material and the diffusion properties (D) of polyhalides, it is essential to consider the significant effects of both porosity and tortuosity (τ) on diffusion paths and mass transfer behavior. When the operational time is much greater than the characteristic diffusion time (t ≫ t_d), the concentration gradient of halogen species in the system gradually stabilizes, and the diffusion process enters a steady state.⁷¹ Under these steady-state diffusion conditions, the relationship between porosity and the polyhalide diffusion rate can be analyzed using Fick's first law: J = −D_e∂C/∂t = −D₀(ε/τ)∂C/∂t,⁷² where the diffusion flux, J, is measured in mol m⁻² s⁻¹; D_e represents the effective diffusion coefficient (m² s⁻¹), indicating the actual diffusion rate within the porous medium; D₀ is the diffusion coefficient in the bulk solution (bulk-phase diffusion); and ∂C/∂t is the rate of change of concentration concerning time (mol m⁻³ s⁻¹). When there is a sudden change in current (e.g., during charge/discharge switching or load variation), the halogen concentration gradient at the electrode/electrolyte interface undergoes dynamic changes.⁷³ At this point, the relationship between porosity and the multi-halide diffusion rate falls under unsteady-state diffusion, which must be described by Fick's second law.⁷⁴ The unsteady-state diffusion control equation is: ∂C/∂t = D_e(∂C/∂t)² = D₀(ε/τ)(∂C/∂t)². The corresponding characteristic diffusion time is t_d ≈ L²/(π²D_e).⁷⁵ In conclusion, the mass transfer behavior of halogen species in zinc–halogen batteries is not only influenced by the coupling control of material structure (such as porosity and tortuosity) and interface dynamics but also by the dominant diffusion mechanism depending on the electrode's working state. Therefore, to enhance the controlled diffusion and efficient conversion of halogen species in energy storage processes, it is essential to design host materials with a confinement–catalysis–conduction structural design, enabling the wide adaptability and regulation of various halogen systems. The following section will systematically review the structural types, mechanisms, and applications of typical halogen-confining host materials in different zinc–halogen batteries.

3.1. Halogen-confining host materials for Zn–I₂ batteries

Zn–I₂ batteries are promising energy storage technologies due to their high intrinsic safety, low cost, and environmentally friendly characteristics. However, they still suffer from several challenges, such as the inherent polyiodide shuttle effect and sluggish reaction kinetics at the iodine cathode, which lead to poor cycling stability, rapid capacity decay, and limited rate performance. Additionally, elemental iodine is inherently insulating, and its reversible redox conversion relies on external conductive carriers to enable efficient electron transfer. To address these issues, various materials including carbonaceous materials (e.g., carbon fiber cloth,⁷⁶ carbon nanotubes/sheets/shells,⁷⁷ graphene,⁷⁸ biomass-derived carbon,⁷⁹ and heteroatom-doped carbon),⁸⁰ functional frameworks (e.g., metal–organic frameworks (MOFs),⁸¹ covalent–organic frameworks (COFs),⁸² covalent organic cages (COCs),⁸³ and PBAs),⁸⁴ derivatives of functional frameworks,⁸⁵ 2D transition metal carbides/nitrides,⁸⁶ and other materials have been developed as conductive hosts for iodine species. These host materials effectively immobilize iodine species through physical confinement within porous structures or chemical adsorption onto active sites, alleviating the dissolution and shuttling effects of polyiodides and improving the electrochemical performance of Zn–I₂ battery systems. This section provides an overview of representative host materials and their confinement strategies for iodine species in Zn–I₂ batteries.

3.1.1. Carbonaceous materials. Carbonaceous materials (e.g., graphene, activated carbon and carbon nanotubes) have been widely explored as host materials for iodine in Zn–I₂ batteries to enhance stability and accelerate electron and ion transport.⁸⁷ These materials offer a robust framework, high specific surface area, excellent conductivity, abundant porosity, and superior chemical and thermal stability.⁸⁸ In particular, porous carbon can physically confine iodine species, trapping iodine within the porous structure and preventing its diffusion into the electrolyte.⁸⁹ This helps mitigate the polyiodide shuttle effect and enhances the utilization of iodine active species during charge and discharge cycles. Moreover, the porous structure provides abundant pathways for ion and electron transport, which further promotes the kinetics of redox reactions.⁹⁰ For instance, Pan et al.⁹¹ developed a highly reversible Zn–I₂ battery (Fig. 7a) by confining iodine within mesoporous carbon as the cathode material. A high iodine loading of up to 54 wt% was achieved by immersing activated carbon fibers (ACF) in an iodine aqueous solution. Scanning electron microscopy (SEM) images (Fig. 7b) show a smooth ACF surface, indicating that iodine is effectively encapsulated within the internal structure of ACF. Fig. 7c demonstrates that after 241 cycles, the Zn–I₂ battery, with a half-charged state of 0.389 mAh cm⁻², maintained a stable open-circuit voltage with only a 10 mV drop over 48 hours of rest, and the discharge capacity after the 242nd cycle showed minimal degradation. This suggests exceptional cycling stability and very low self-discharge. The good electrochemical performance is attributed to the mesoporous structure of the ACF host, which provides stable iodine storage, significantly suppresses the dissolution and shuttle effects of I₃⁻, and ensures continuous redox reaction sites during charge/discharge processes. DFT simulations indicate that in aqueous environments, I₂ and Zn(I₃)₂ preferentially adsorb onto the carbon host surface instead of dissolving, effectively preventing self-discharge (Fig. 7d and e). In contrast, in organic solvents, both species exhibit positive interaction energy difference (ΔE) values, indicating a tendency for dissolution and poor adsorption, leading to enhanced shuttle effects (Fig. 7f and g). Additionally, Fig. 7h shows that the pore size of the carbon material plays a crucial role in the confinement capacity: larger pores weaken iodine adsorption, diminishing confinement effects and, consequently, reducing battery capacity and CE under the same iodine loading conditions. These experimental and theoretical findings validate that stable confinement of iodine ions within mesoporous ACF is essential for achieving ultra-stable cycling performance and high CE in aqueous Zn–I₂ batteries. Building on the critical role of mesoporous carbon hosts in iodine confinement, subsequent studies have proposed multidimensional synergistic confinement strategies. These strategies include the development of carbon-based hosts with higher surface areas and optimized pore size distributions, along with the incorporation of functionalized separators to further enhance iodine species fixation and mitigate shuttle behavior. For example, Yu et al.⁹² proposed using mesoporous carbon nanotubes (MCN) in combination with modified cotton fiber separators (MCN@CF) to create a confinement system, which was used to address the high solubility of iodine species (Fig. 7i). MCN exhibits an exceptionally high specific surface area and a pore size of approximately 1.93 nm (Fig. 7j), offering abundant active sites for iodine adsorption and stabilization. This significantly reduces iodine species migration to the anode and suppresses the shuttle effect. Batteries assembled with MCN/I₂ cathodes and MCN@CF separators exhibited high specific capacity and excellent capacity retention. This research confirms that mesoporous carbon-based hosts can effectively mitigate iodine shuttle behavior through stable adsorption, and the synergistic combination of functionalized separators further enhances the cycling stability and energy density of Zn–I₂ batteries. In recent years, various biomass materials such as leaves, corncobs, seaweed, and walnut shells have been widely used to develop porous carbonaceous materials.⁹³ These biomass-derived materials retain their natural hierarchical structures and transform into porous carbon structures with high specific surface areas and tunable pore sizes through carbonization and chemical activation (e.g., KOH or ZnCl₂).⁹⁴ These structures provide abundant active sites for the adsorption and immobilization of iodine species, significantly mitigating iodine diffusion and shuttle effects in Zn–I₂ batteries. Additionally, certain biomass materials naturally retain heteroatoms (e.g., N, S and O) during carbonization, forming heteroatom-doped carbon, which enhances iodine binding and electrochemical reactivity.⁹⁵ This offers an effective pathway to fabricate high-performance, low-cost Zn–I₂ batteries. For example, Wei et al.⁹⁶ developed an oxygen-doped mesoporous carbon material derived from cotton (OHPCF), which serves as a host for iodine species in Zn–I₂ batteries. The synthesized OHPCF-0.5 features a high-density microporous structure and significant oxygen doping content. Raman spectroscopy analysis revealed that no characteristic iodine peaks were observed after iodine adsorption onto OHPCF, and the ID/IG ratio remained largely unchanged, further confirming that iodine is effectively confined in an amorphous state within the carbon framework (Fig. 7k). DFT calculations and charge density difference analysis indicated that oxygen atoms can attract electrons from adjacent carbon atoms, inducing local carbon atoms to adopt a positively charged environment, which enhances their chemical adsorption capacity for iodine (Fig. 7l). This charge polarization effect not only strengthens the confinement of iodine within the carbon structure but also emphasizes the critical role of oxygen doping in stabilizing iodine species. Thanks to the combined physical confinement effect of the high-density microporous structure and the chemical anchoring effect of oxygen doping, the OHPCF electrode exhibits low polarization, high iodine utilization efficiency, and excellent electrochemical stability. These findings confirm the pivotal role of microporous structures in regulating iodine species behavior, where spatial confinement partially suppresses the diffusion and shuttle effects of iodine species. Moreover, the pore size characteristics and surface chemistry directly improve iodine adsorption strength and confinement efficiency, which thus enhance the electrochemical performance of Zn–I₂ batteries.

Fig. 7 (a) Schematic of the aqueous Zn–I₂ battery with iodine confined in a microporous carbon as a cathode. (b) SEM image of the 40I₂/ACF before cycling. (c) Cycling performance of the Zn–I₂/ACF battery after resting at half-charge for 48 h (241st cycle). ΔE between the dissolution of I₂ and Zn(I₃)₂ in water (d) and (e) and propylene carbonate solvents (f) and (g) and their corresponding adsorption energy on the carbon surface. (h) Maximum areal iodine loading, capacity, and CE of Zn–I₂ batteries on carbon matrices with different pore sizes.⁹¹ Reproduced with permission from ref. 91. Copyright 2017, American Chemical Society. (i) Schematic of a Zn–I₂ battery assembled with MCN/I₂ as a cathode and MCN@CF as a separator. (j) SEM image of MCN.⁹² Reproduced with permission from ref. 92. Copyright 2024, Elsevier. (k) Raman spectra of OHPCF-0.2, OHPCF-0.5 and OHPCF-1.0. (l) Schematic of the optimized configuration of OHPCF-0.5, along with its charge density difference in both the overall structure and the XY plane.⁹⁶ Reproduced with permission from ref. 96. Copyright 2024, Elsevier.

Traditional Zn–I₂ systems utilizing I₂ as the active material often suffer from severe volume expansion during the initial zincation process, leading to structural breakdown and performance degradation (Fig. 8a), which serves as a major barrier to their further development. To mitigate the structural instability caused by these volume effects while ensuring efficient fixation and electrochemical conversion of the active material, researchers have explored the direct use of fully zincated ZnI₂ as the cathode active material (Fig. 8b), incorporating functionalized mesoporous carbon host materials to synergistically optimize electrode performance. Ma et al.⁹⁷ replaced I₂ with ZnI₂ as the cathode active material and loaded it onto carbon materials modified with highly active cobalt nanocrystals, resulting in a composite cathode material (NC–Co) with both microporous and mesoporous structures. This hierarchical pore structure provides ample space for high ZnI₂ loading and buffers a volume shrinkage of up to 23.52% during de-zincation, significantly enhancing the structural stability of the electrode. Compared to the N-doped carbon (NC) host, NC–Co substantially improves the confinement and electrocatalytic conversion of iodine intermediates, thereby enhancing the specific capacity of ZnI₂ (Fig. 8c and d). In situ Raman tests further confirm the chemical anchoring effect of NC–Co on polyiodide anions and its stability during the reaction process (Fig. 8e and f). During charging, the characteristic peak intensities of I₃⁻ and I₅⁻ in the NC–Co/ZnI₂ electrode gradually increase as the voltage rises, reaching a maximum at 1.6 V; while during discharge, these peaks gradually weaken and eventually disappear around 0.4 V, indicating that iodine species undergo good reversible I⁻ ↔ I₃⁻ ↔ I₅⁻ conversion. In contrast, the iodine intermediates in the NC/ZnI₂ electrode show incomplete conversion, slower reaction kinetics, and reduced reversibility and utilization of iodine species (Fig. 8g). Thanks to the pore structure design and the synergistic effect between nitrogen-doped carbon atoms and cobalt nanocrystals in the porous matrix, the NC–Co/ZnI₂ electrode exhibits excellent initial capacity and retains a high reversible capacity after 2000 cycles, with a significant advantage in energy density and specific capacity compared to other materials used in Zn–I₂ batteries (Fig. 8h). This study ingeniously combines fully zincated active materials with multifunctional mesoporous carbon hosts, achieving dual synergistic effects of structural buffering and interface catalysis, balancing both structural stability and efficient conversion and confinement of iodine species. This not only overcomes the structural degradation bottleneck of traditional I₂ cathodes but also provides a viable strategy for constructing high-load, high-cycle-stability Zn–I₂ batteries. To further enhance the reaction kinetics of iodine species and suppress iodine sublimation and shuttle effects under high-temperature conditions, researchers have focused on heterostructure engineering and the construction of atomic-level catalytic sites in host materials to improve the electrocatalytic conversion efficiency of iodine. Recently, Chen et al.⁹⁸ employed a templating method combined with molten salt-assisted strategies to design and construct a hierarchical porous carbon material (PC@Fe₂N) loaded with highly dispersed Fe₂N clusters, which serves as a host material for iodine cathodes. The hierarchical mesoporous structure of this material provides ample storage space for high iodine loading, while the highly dispersed Fe₂N clusters significantly improve the surface charge characteristics of the carbon substrate, thereby enhancing the iodine catalytic conversion rate and effectively alleviating iodine sublimation and shuttle effects under high-temperature conditions. To gain deeper insights into the electrode reaction process, in situ Raman spectroscopy was used to monitor the charging and discharging processes (Fig. 8i and j). During discharge, the intensity of the I₃⁻ peak gradually weakens, indicating its conversion to I₂. The simultaneous weakening of the I₅⁻ and I₃⁻ signals indicates a shift in the multi-iodine anions towards a direct I₂ ↔ I⁻ reaction pathway, showcasing excellent electrocatalytic activity favorable for reversible reactions. In contrast, in the comparison material PC@Fe₂N-0/I₂, distinct I₃⁻ and I₅⁻ peaks are still visible, reflecting slower iodine conversion kinetics and poorer electrocatalytic performance. Further analysis of the projected density of states (PDOS) revealed that the orbital hybridization between Fe₂N and iodine effectively modulates the electronic structure of the carbon substrate, lowering the reaction energy barrier (Fig. 8k). Differential charge density and Mulliken charge distribution results show that the N-graphene-Fe₂N clusters exhibit stronger polarity, and significant charge redistribution occurs on the surface when in contact with iodine, demonstrating enhanced electronic migration ability (Fig. 8l), which effectively facilitates the redox conversion of iodine species and improves the reaction rate and electrochemical stability. The corresponding reaction pathway diagram (Fig. 8m) further validates its efficient and reversible electrochemical behavior. Ultimately, the constructed PC@Fe₂N structure in Zn–I₂ batteries with high iodine loading maintains a reversible capacity of 148 mAh g⁻¹ at 60 °C and demonstrates an ultra-long cycle life of over 20 000 cycles (Fig. 8n). This study, through the construction of PC@Fe₂N carbon hosts with high-polarity catalytic sites and hierarchical pore structures, significantly enhances both the electrocatalytic conversion efficiency and cycling stability of iodine species, providing an effective and efficient material design strategy for Zn–I₂ batteries capable of stable operation at elevated temperatures.

Fig. 8 (a) Schematic of the structural evolution of the material with iodine and ZnI₂ as active species upon the charge/discharge process. (b) Comparison of energy density between the thick ZnI₂ electrode and the conventional thin I₂ electrode. (c) Voltage-specific capacity curves and (d) specific capacities of AC/ZnI₂, NC–Co/I₂, NC/ZnI₂, and NC–Co/ZnI₂ electrodes, respectively. In situ Raman spectra of (e) NC–Co/ZnI₂ and (f) NC/ZnI₂ electrodes during various stages of the charge–discharge process. (g) Comparison of iodine conversion kinetics in the NC/ZnI₂ electrode and NC–Co/ZnI₂ electrode. (h) Comparison of the electrochemical performance of the iodine cathode with reports in previous studies.⁹⁷ Reproduced with permission from ref. 97. Copyright 2024, Royal Society of Chemistry. In situ Raman spectra of (i) PC@Fe₂N-4/I₂ and (j) PC@Fe₂N-0/I₂ electrodes during the charge–discharge process. (k) Schematic of the PDOS changes of N-graphene-Fe₂N and N-graphene-Fe₃C before and after I₂ adsorption. (l) Schematic of the charge density differences of I₂ on graphene, N-graphene, N-graphene-Fe₂N, and N-graphene-Fe₃C. (m) ΔG during the halogen reduction reaction (IRR) on graphene, N-graphene, N-graphene-Fe₂N, and N-graphene-Fe₃C substrates. (n) Cycling performance of PC@Fe₂N-4/I₂ at a current density of 10C, with the inset showing the corresponding charge–discharge curves.⁹⁸ Reproduced with permission from ref. 98. Copyright 2023, Elsevier.

Recently, researchers have extensively explored various transition metal SACs, but due to the diversity of d-block metals, the construction rules for their optimal active centers remain unclear.⁹⁹ Previous studies have primarily focused on calculating adsorption energies and reaction energies during discharge, while the understanding of the relationship between the d-electronic structure and catalytic performance in SACs is still insufficient. However, the d-orbital structure of the metal center determines the chemical bonding and reactivity with iodine species. Based on this, Yang et al.¹⁰⁰ systematically studied the regulation of polyiodide anion conversion by 12 typical d-block transition metals in M–N₄ coordination structures using DFT calculations (Fig. 9a). The results indicate that the main interaction between iodine species and the metal active center arises from d–p orbital hybridization, and the arrangement of d orbital energy levels directly affects the hybridization efficiency. For example, due to the relatively low number of d-electrons in Nb atoms, the d orbital energy levels redistribute, causing the d_xz/yz levels to shift upward, while the d_xy and d_x²−y² levels shift downward. This makes bonding orbitals more readily occupied, forming unfilled anti-bonding states and bringing the d-band center closer to the Fermi level (0.271 eV), which significantly enhances the d–p hybridization strength with iodine species (Fig. 9b). This structure not only gives it a strong adsorption capacity for iodine species but also exhibits the lowest reaction energy barrier in the key oxidation–reduction step (I₃⁻ → I⁻), showing excellent kinetic characteristics. Inspired by this theory, the authors further synthesized Nb-metal-free nitrogen-doped carbon (Nb–NC) materials and constructed them as a self-supporting I₂ electrode without a current collector, achieving in situ confinement and efficient conversion of iodine species. This effectively improved the volumetric energy density and cycling stability of Zn–I₂ batteries. DFT calculations showed that all M–NC catalysts exhibited higher adsorption energies for I⁻, I₂, I₃⁻, and I₅⁻ than NC, indicating stronger affinity for iodine species (Fig. 9c). In the density of states (DOS) analysis, the M–NC systems constructed using Nb, Co, Re, Rh, Ru, Ti, W, Zn, and other metals exhibit conduction band (CB) crossing the Fermi level, reflecting good conductivity conducive to charge transfer; in contrast, the NC structure has a wide band gap, which is not favorable for efficient electrochemical behavior (Fig. 9d). Further electronic structure analysis of the SAC–ZnI₂ system revealed that the d-band center of Nb–NC is close to the Fermi level, and it has the strongest adsorption energy for I₂, providing a favorable thermodynamic basis for the reaction (Fig. 9e). In the complete I₂ reduction path (I₂ → I₂* → Zn(I₃)₂* → ZnI₂*), Nb–NC showed the smallest ΔG in the intermediate conversion process, especially in the rate-determining step of I₃⁻ → I⁻, with the lowest reaction energy barrier, further proving the unique advantage of the Nb–N₄ structure in enhancing kinetics (Fig. 9f). In situ UV-Vis spectroscopy showed that the characteristic absorption peak of I₃⁻ gradually weakened during discharge and then gradually recovered during charge, reflecting the reversible conversion behavior of polyiodide anions (Fig. 9g and h). Especially in the Zn‖Nb–NC/I₂ battery system, compared to Zn‖NC/I₂, the absorption peak of I₃⁻ significantly weakened, indicating that Nb–NC has a stronger inhibitory effect on intermediate dissolution and shuttle effects (Fig. 9i). Moreover, its excellent adsorption capacity effectively prevents I₃⁻ migration to the zinc anode, which would otherwise lead to side reactions and corrosion. Thanks to its conductive carbon network and rationally constructed active sites, the self-supporting Nb–NC/I₂ electrode demonstrates excellent rate performance and cycling stability. This system has been validated for its significant potential in achieving iodine species confinement and accelerating electrocatalytic conversion.

Fig. 9 (a) Adsorption and catalytic behavior of M–NC (M = Ti, Fe, Co, Ni, Cu, Zn, Nb, Mo, Ru, Rh, W, and Re) on iodine species. (b) Illustration of the d–p orbital hybridization scenario. (c) Schematic of the adsorption energy of I⁻, I₂, I₃⁻, and I₅⁻ on M–NC and NC. (d) DOS plots of M–NC (M = Nb, Mo, Fe, Co, Cu) and NC. (e) Correlation between the d-band center of various SACs and their adsorption energy for I₂*. (f) Gibbs free energy profiles for iodine reduction on M–NC and NC. (g) In situ UV-vis experimental setup diagram. (h) In situ UV-vis spectra of the Nb-NC/I₂ electrode at various charge and discharge states. (i) Adsorption and catalytic behavior of Nb–NC on polyiodide species.¹⁰⁰ Reproduced with permission from ref. 100. Copyright 2025, Royal Society of Chemistry.

Although the microporous/mesoporous structures in porous carbon materials can physically adsorb iodine species to some extent, their ability to confine iodine species is limited due to the weak van der Waals forces between the nonpolar carbon matrix and polar iodine molecules.⁸⁹ This leads to ineffective suppression of polyiodide anions’ dissolution and diffusion, causing shuttle behavior between the cathode and anode, resulting in active material loss, increased self-discharge, and reduced cycling stability. Compared to the physical adsorption in nonpolar carbon frameworks, combining carbon materials with polymers containing polar functional groups such as hydroxyl and amine groups can enhance iodine species confinement through both physical and chemical interactions, effectively suppressing their dissolution and diffusion.¹⁰¹ For example, Wei et al.¹⁰² employed a simple soft-hard template co-assembly strategy to in situ polymerize the aniline derivative 1,8-diaminonaphthalene (DAN) onto graphene oxide (GO) nanosheets, successfully preparing ordered mesoporous poly-1,8-diaminonaphthalene layers (PDAN) grown on reduced graphene oxide (rGO) surfaces to form conductive nanosheets (mPD@rGO) (Fig. 10a). mPD@rGO exhibits regular micron-sized nanosheets with uniformly distributed, densely packed, ordered mesopores on both the front and back surfaces (Fig. 10b and c). These mesopores are arranged in a short-range ordered hexagonal pattern and partially overlap between nanosheet layers, indicating that the pore structure is distributed at different levels, forming a multi-level cooperative confinement network that effectively enhances iodine species confinement (Fig. 10d). By loading iodine onto mPD@rGO and rGO using sublimation diffusion, the resulting mPD@rGO/I₂ exhibited an iodine adsorption capacity of up to 52.4 wt%, demonstrating excellent iodine confinement capability. In adsorption performance tests, mPD@rGO was able to almost completely remove I₃⁻ from the solution in a short period, showing stronger polyiodide anion confinement compared to rGO (Fig. 10e). To further investigate the adsorption mechanism of iodine species on the composite electrode, DFT calculations revealed that both PDAN and rGO exhibit thermodynamically spontaneous adsorption for I₂ and I₃⁻, with PDAN showing a much higher adsorption energy for I₃⁻ than rGO, indicating that the polymer component plays a dominant role in confining polyiodide anions (Fig. 10f and g). Additionally, in situ XRD revealed structural evolution in mPD@rGO during the charge and discharge process. A new peak appeared at 7.9° during discharge, indicating the formation of ZnI₂, and the peak disappeared after full charge, confirming the reversible conversion of iodine species (Fig. 10h). This excellent reversibility and rapid reaction help suppress the shuttle effect of polyiodide anions, improving the utilization of active materials. Furthermore, the study found that for the highly polar I₃⁻, not only does electron transfer occur at the nitrogen sites of PDAN, but it also forms synergistic charge interactions with the surrounding carbon atoms, indicating that both PDAN and rGO have advantages in confining I₃⁻ (Fig. 10i). The PDOS calculations showed that in the rGO–I₃⁻ system, the PDOS of iodine atoms is widely distributed near the Fermi level and overlaps with C orbitals, while in the PDAN–I₃⁻ system, more significant orbital hybridization between I and N is observed (Fig. 10j). This synergistic confinement mechanism not only effectively suppresses the shuttle behavior of I₃⁻ but also promotes the reversibility and electron transfer of electrode reactions, ensuring the long-term stable operation of Zn–I₂ batteries. Therefore, the iodine-loaded mPD@rGO cathode material in the Zn–I₂ battery demonstrates a specific capacity of up to 271.4 mAh g⁻¹, excellent rate performance, and good long-term cycling stability. This work reveals that constructing polymer–carbon composite nanosheets with a multi-level mesoporous structure and synergistic polar functional group interactions not only significantly enhances iodine species confinement and adsorption but also effectively promotes their electrochemical reversible conversion.

Fig. 10 (a) Schematic diagram of the synthesis of mPD@rGO. (b) SEM image, (c) atomic force microscopy (AFM) image with the corresponding height curve shown in the inset, (d) transmission electron microscopy (TEM) image of mPD@rGO. (e) UV-Vis spectra of rGO and mPD@rGO soaked in I₃⁻ solutions for 2 h, with digital images shown in the inset. (f) Adsorption energy and (g) structure models of rGO and PDAN for absorption of I₂ and I₃⁻. (h) Ex situ XRD patterns of the mPD@rGO/I₂ electrode at various charge and discharge states. (i) Charge density models of I₃⁻ on rGO and PDAN. (j) Total DOS (TDOS) and PDOS plots of GO-I₃⁻ and PDAN-I₃⁻.¹⁰² Reproduced with permission from ref. 102. Copyright 2023, Wiley-VCH.

In summary, carbonaceous materials have shown significant potential as halogen-confining host materials in Zn–I₂ batteries, enhancing both specific capacity and cycling efficiency. However, their performance is often constrained by limited halogen species confinement and unfavorable redox kinetics. Heteroatom doping (e.g., N, O and S) and surface functional group modifications (e.g., hydroxyl and amine groups) impart stronger chemical anchoring abilities and electrocatalytic activity, effectively promoting the reversible conversion of polyiodide anions. The further introduction of metal clusters (such as Fe₂N) or the construction of heterostructures helps to synergistically improve iodine conversion kinetics and electrode stability. Despite significant progress in the adsorption, confinement, and catalytic conversion of iodine species, several challenges remain, and future research should focus on the below: (i) designing carbon-based hosts with tunable pore structures and high electrocatalytic activity to achieve synergistic enhancement of both confinement and catalytic conversion; (ii) constructing an integrated structural system with both volume buffering and interface stability functions, achieving high iodine loading while maintaining structural integrity and reversibility of the reactions; (iii) integrating in situ characterization techniques and multiscale simulations to reveal the reaction mechanisms of iodine species in confinement systems. Additionally, by loading iodine onto flexible mesoporous carbon fibers, self-supporting electrodes can be constructed without the need for binders and conductive agents, which can also be applied in the development of flexible batteries, thus expanding the application range of Zn–I₂ batteries.

3.1.2. Functional frameworks. Functional frameworks, such as MOFs, COFs, and PBAs, exhibit highly ordered pore structures and large specific surface areas.¹⁰³ When these structures are used as halogen confinement host materials in Zn–I₂ batteries, they can effectively enhance the adsorption and electron transfer efficiency of iodine species while reducing the loss of active materials.⁸⁹ Furthermore, these framework structures are tunable, and by introducing functional groups or metal centers, they can enhance the materials’ ability to adsorb and fix iodine species, reduce the dissolution and shuttle effects of iodine species, and thus improve the cycling stability and electrochemical performance of the battery.¹⁰⁴ These characteristics make functional frameworks ideal candidates as cathode host materials for high-performance Zn–I₂ batteries. 2D conjugated metal–organic frameworks (2D c-MOFs), as a novel class of crystalline porous polymers, exhibit high conductivity, tunable pore sizes, and abundant active sites, making them favorable for iodine species adsorption and accelerating iodine conversion reaction kinetics.¹⁰⁵ However, most of the reported 2D c-MOFs are microporous. To address this, Bao et al.¹⁰⁶ successfully synthesized two new types of 2D c-MOFs with mesoporous structures using a simple solvothermal method, based on multitopic catechol ligands (6OH-PA-TAPA and 8OH-PA-PyTTA), namely the PA-TAPA-Cu-MOF (Fig. 11a) and the PA-PyTTA-Cu-MOF. The entire nanocrystals of the PA-TAPA-Cu-MOF exhibit a honeycomb-like pore arrangement (Fig. 11b and c), indicating that this MOF possesses an ordered, through-hole mesoporous structure, which is conducive to iodine species confinement and transport. Molecular electrostatic potential (MESP) comparisons reveal that, compared to the PA-PyTTA-Cu-MOF, the PA-TAPA-Cu-MOF shows more abundant active sites in the ligand center region, including nitrogen atoms from triphenylamine, nitrogen atoms from aromatic imides, and the [CuO₄] coordination centers (Fig. 11d). Further adsorption energy calculations show that all three active sites exhibit negative adsorption energies when interacting with I⁻, I₃⁻, and I₂, indicating that they possess good spontaneous adsorption capacity for iodine and its intermediates, facilitating efficient iodine species confinement (Fig. 11e). Additionally, a comparison of the reduction pathway ΔG at the three sites reveals that site 3 ([CuO₄]) has the lowest ΔG, suggesting that it plays a dominant role in the confinement of I₃⁻ (Fig. 11f). These results highlight the effectiveness of constructing ordered mesoporous MOF structures with metal coordination centers, which can significantly enhance the stable adsorption and suppression of iodine species shuttle behavior.

Fig. 11 (a) Schematic diagram of the preparation process for the PA-TAPA-Cu-MOF. (b) High-resolution TEM (HRTEM) image of the PA-TAPA-Cu-MOF. (c) Magnified image corresponding to the red box region in (b). (d) MESP of repeat units for the PA-PyTTA-Cu-MOF and the PA-TAPA-Cu-MOF. (e) Possible active sites of the PA-TAPA-Cu-MOF and the adsorption energy of I₂, I₃⁻, and I⁻ at different active sites. (f) Gibbs free energy profiles of the I₂ reduction reaction at different active sites of the PA-TAPA-Cu-MOF.¹⁰⁶ Reproduced with permission from ref. 106. Copyright 2024, Wiley-VCH. (g) Galvanostatic charge–discharge (GCD) curve of the IL-ZIF-90-I⁻ electrode and its in situ UV-Vis spectra after 1000 cycles. (h) Optimized charge density difference model for I⁻, I₂, and I₃⁻ adsorbed at C-site N. (i) GCD curves of the IL-ZIF-90-I⁻ electrode at different current densities.¹⁰⁷ Reproduced with permission from ref. 107. Copyright 2024, Royal Society of Chemistry. (j) Structural schematic of MFM-300(Sc) projected along the [001] direction. HAADF-STEM (k) and iDPC-STEM (l) images of pristine MFM-300(Sc). (m) Iodine uptake of MFM-300(Sc) over time under saturated I₂ vapor at 75 °C.¹⁰⁸ Reproduced with permission from ref. 108. Copyright 2025, American Chemical Society.

In addition to utilizing materials with pore structures to confine iodine species, researchers have also attempted to enhance their confinement ability and suppress shuttle effects by functionalizing and modifying the MOFs. He et al.¹⁰⁷ successfully synthesized iodine-covalently functionalized zeolitic imidazolate framework-90 (IL-ZIF-90), which formed multifunctional nitrogen sites inside the material, enabling strong adsorption of iodine species. In situ UV-vis absorption spectroscopy results indicated that in this confined environment, the redox process primarily involves the direct conversion between I⁻ and I₂, avoiding the formation of polyiodide anions (I₃⁻) and effectively suppressing the shuttle effect (Fig. 11g). Further calculations revealed that interactions between the nitrogen atoms at positions C/D and iodine species could securely confine iodine species (Fig. 11h). Moreover, as the current density increased, IL-ZIF-90-I⁻ consistently exhibited stable charging and discharging platforms (Fig. 11i). This dual interaction mechanism between the nitrogen functional sites and iodine gives IL-ZIF-90-I⁻ a significant advantage in stabilizing iodine species, preventing active material loss and maintaining reaction reversibility, offering an important approach for the design of new, efficient iodine host materials. To further elucidate the iodine species confinement mechanism in MOF materials, Liu et al.¹⁰⁸ visualized the adsorption behavior of I₂ in the MOF material MFM-300(Sc) using high-resolution scanning TEM (STEM). The results showed that the synthesized MFM-300 has 1D channels (Fig. 11j), and the channels in MFM-300(Sc) are empty, providing ample space for iodine species adsorption (Fig. 11k and l). Adsorption experiments were conducted in a saturated I₂ vapor environment, and iodine absorption was tracked by measuring mass changes. The sample achieved an iodine adsorption capacity of 1.47 g g⁻¹ within 10 hours, consistent with the saturated adsorption capacity reported in the literature (Fig. 11m). More importantly, the study found that the confinement adsorption of iodine molecules is heterogeneous even in the saturated adsorption state. This provides key evidence for understanding the confinement regulation mechanism of iodine species in MOFs at the atomic scale. The aforementioned work demonstrates that MOFs, with their high designability, multifunctional coordination sites, and ordered pore structures, exhibit significant advantages as iodine species confinement host materials in Zn–I₂ batteries. Furthermore, the integration of in situ characterization and theoretical calculations has provided deep insights into the distribution behavior of iodine within MOF channels and the host–guest interaction mechanisms, offering crucial evidence for understanding the confinement process at the molecular and atomic scales.

COFs often offer the advantages of high crystallinity and large porosity, making them promising candidates as iodine species hosts for Zn–I₂ batteries.¹⁰⁹ However, conventional COF synthesis methods can result in unpredictable nucleation, growth, and aggregation of crystals, leading to arbitrary orientations and geometries.¹¹⁰ Inspired by the biosynthesis of natural products in nature, Li et al.¹¹¹ developed a supramolecular mineralization (SM) strategy, which controls the pre-assembly of supramolecules and adjusts the rigid connections between organic molecules to reduce the entropy of host materials, thereby altering the dynamic covalent chemical reaction pathways and achieving the controlled synthesis of COF nanostructures (Fig. 12a and b). Using this method, three COFs with specific hollow nanostructures and 2D structures were successfully synthesized by matching different organic molecules, named PY-1P, PY-2P, and PY-2PBA (Fig. 12c). DFT results indicated that the nitrogen atoms in the bipyridine structural units serve as electron-withdrawing sites, lowering the energy of the lowest unoccupied molecular orbital (LUMO) and enhancing the electron affinity (Fig. 12d). Moreover, the electrostatic potential (ESP) of the PY-2PBA COF (Fig. 12e) visually illustrates the surface potential distribution, which helps predict the active sites for iodine confinement. Adsorption experiments further revealed that among the synthesized COFs, PY-2PBA showed the highest iodine adsorption capacity, with DFT calculations providing additional quantitative evidence (Fig. 12f). Furthermore, because I₂ forms a charge-transfer complex with the PY-2PBA COF, its conductivity significantly increases after iodine adsorption, making it more advantageous as an iodine host material for Zn–I₂ batteries. The iodine-loaded I₂@PY-2PBA/carbon nanotube cathode demonstrated high specific capacity and excellent long-term cycling stability. This study, through the synergistic regulation of non-covalent and dynamic covalent interactions, combines supramolecular chemistry with polymers, not only achieving the controllable construction of COFs and supramolecular materials but also providing a new approach for the development of customized COFs and self-assembled materials, expanding their application potential in various fields. However, the simultaneous regulation of the mesostructure and configuration of COFs is crucial for their applications. Hierarchical porous COFs, with rich active sites, fast ion diffusion channels, and anisotropic structural advantages, show broad application prospects in electrochemical energy storage. Based on this, Huang et al.¹¹² proposed a polymerization-induced co-assembly (PICA) strategy and constructed a 2D hierarchical porous COF material with both intrinsic microporosity and tunable macroporosity under aqueous conditions (HPCOF@GO-x, where x represents the particle size of the SiO₂ template) (Fig. 12g). The TEM image of HPCOF@GO-22 showed uniformly distributed large mesopores densely packed on its surface (Fig. 12h), and AFM further confirmed this, revealing its smooth surface and uniform thickness (Fig. 12i). Given the unique 2D structure and the excellent physical–chemical adsorption of iodine provided by the fully exposed hierarchical pores, adsorption experiments and DFT calculations were performed (Fig. 12j). The results confirmed that HPCOF@GO-22 nanosheets exhibit strong adsorption interactions with polyiodide species, effectively suppressing the shuttle effect and accelerating the conversion of polyiodide anions. Therefore, as an iodine host material in aqueous Zn–I₂ batteries, it exhibited satisfactory rate performance (Fig. 12k) and maintained good cycling stability at a current density of 3 A g⁻¹ (Fig. 12l). This work demonstrates that hierarchical pore channels not only effectively confine iodine species but also accelerate the conversion of polyiodide anions, highlighting the important role of hierarchical pore structures in iodine host material design. These studies demonstrate the feasibility of constructing high-performance COF host materials through structural programmability and multi-scale pore channel regulation. Future research on iodine-confining COF host materials should place greater emphasis on structural programmability and multifunctional integration, utilizing AI-assisted molecular design and high-throughput simulations to explore optimal combinations of structural units. The goal is to develop iodine host materials with excellent conductivity, strong confinement capabilities, and fast reaction kinetics, thereby guiding the rational design of halogen species host materials for high-energy-density, long-lifetime, and low self-discharge Zn–I₂ batteries.

Fig. 12 (a) Traditional route and SM route for the synthesis of COFs. (b) Energy changes between the traditional route and the SM route during COF construction. (c) Synthesis and structural diagrams of PY-1P, PY-2P, and PY-2BPA COFs. (d) The highest occupied molecular orbital (HOMO) and LUMO of the PY-1P, PY-2P, and PY-2BPA COFs. (e) ESP of the PY-2BPA COFs. (f) Adsorption energy of I⁻, I₂, and I₃⁻ species on the PY-2BPA COF.¹¹¹ Reproduced with permission from ref. 111. Copyright 2023, Elsevier. (g) Schematic of the fabrication process of HPCOF@GO. (h) TEM image, (i) AFM image of HPCOF@GO-22. (j) Adsorption energy of I₃⁻, I₂, and I⁻ on HPCOF@GO-22. (k) Rate capability of I₂@HPCOF@GO-22 and I₂@COF@GO electrodes at various current densities. (l) Long-term cycling performance of the I₂@HPCOF@GO-22 electrode.¹¹² Reproduced with permission from ref. 112. Copyright 2024, Elsevier.

To further enhance the iodine confinement ability and reaction activity of COFs in Zn–I₂ batteries, precise molecular-level control over their electronic structure and active site distribution is required. Recently, Yin et al.¹⁰⁹ employed a molecular engineering strategy by selecting amino monomers with different nitrogen densities to construct a series of N_x-COF materials (x = 1, 2, 3, 4) with similar pore structures but varying nitrogen content and distribution (Fig. 13a). These materials were systematically applied as iodine host materials in Zn–I₂ batteries (Fig. 13b). Experimental results showed that as the nitrogen content increased, the internal resistance of I₂@N_x-COF gradually decreased, and its redox reaction kinetics were significantly improved. Fig. 13c displays the four types of nitrogen sites in I₂@N₄-COF, and calculation results indicate (Fig. 13d) that all four sites exhibit negative adsorption energy for I₃⁻, with site 4 showing the strongest binding strength for iodine. The differential charge map shows the charge transfer from N₄-COF to I₃⁻, further confirming the strong interaction between site 4 and iodine (Fig. 13e). The DOS analysis of iodine species in I₂@N₄-COF (Fig. 13f) shows that the iodine anchored in the N₄-COF exhibits a narrower bandgap, indicating a stronger electron-accepting ability, which effectively enhances reaction kinetics, and the COF structure facilitates efficient electron transfer for iodine species. Furthermore, the TDOS of the N₄-COF (Fig. 13g) shows higher electron density near the valence band (VB), suggesting better conductivity. Among this series of materials, the nitrogen-rich N₄-COF can effectively confine iodine species, alleviate shuttle effects, and exhibit the fastest iodine conversion kinetics and the highest electrochemical activity. Therefore, the I₂@N₄-COF-based Zn–I₂ battery achieved a specific capacity of up to 348 mAh g⁻¹ and excellent cycling stability. This work systematically reveals the impact of nitrogen content and distribution on COF materials’ ability to confine iodine species, charge transfer behavior, and reaction kinetics, demonstrating the great potential of nitrogen-containing COFs as efficient iodine host materials in Zn–I₂ batteries. Given that doping variable-valence metal centers and introducing adsorption sites for polyiodide species can significantly accelerate the redox kinetics of I⁻/I₂ or I⁻/I₃⁻ pairs and effectively suppress polyiodide species shuttle, Feng et al.¹¹³ reported a strategy that combines metal active centers with functional adsorption sites. They prepared six woven COFs with ruthenium (Ru) redox centers and sulfur (S) adsorption sites, namely COF-RuNCS-X (X = 1–6) (Fig. 13h). These materials combine structural order with pore channel regulation capabilities, enabling iodine species confinement. DFT calculations revealed that the adsorption energies of I₃⁻ and I₅⁻ polyiodide anions on the RuNCS sites are lower than that of I⁻ (Fig. 13i), indicating that this structure can selectively strongly adsorb more shuttle-prone iodine intermediates, effectively suppressing their shuttle behavior and stabilizing their spatial distribution. Furthermore, the redox activity of the Ru metal centers can significantly promote the electron transfer process between I⁻/I₂ and I⁻/I₃⁻, synergistically enhancing iodine conversion kinetics. The charge–discharge mechanism of the Zn–I₂ battery based on COF-RuNCS-X is shown in Fig. 13j. When the Zn–I₂ battery is connected to an external circuit, Ru can inject electrons into the COF-RuNCS-X framework, after which electrons flow into the zinc anode through the external circuit and then transfer to I₃⁻ and I₅⁻via the electrolyte, reducing them to I⁻. Finally, I⁻ transfers electrons back to Ru, completing a full charge–discharge cycle. Notably, among the six COFs prepared, COF-RuNCS-6, which has a larger pore size and higher conjugation, exhibited stronger iodine confinement. When used as a cathode host material in Zn–I₂ batteries, it achieved a high specific capacity of 395.8 mAh g⁻¹ and excellent cycling stability for up to 5000 cycles. This demonstrates that the introduction of Ru-NCS functional sites not only provides excellent redox activity but also enhances selective adsorption of intermediates, significantly improving iodine conversion kinetics and cycling stability. These findings highlight the critical role of fine-tuning molecular structures to enhance COF confinement performance through electronic structure and functional site engineering, offering valuable insights for the development of efficient and stable iodine host materials in high-performance Zn–I₂ batteries.

Fig. 13 (a) Schematic diagram of the synthesis of the N_x-COF and the corresponding molecular structures. (b) Advantages of the N_x-COF as a host material for Zn–I₂ batteries. (c) Possible active nitrogen sites of the N₄-COF. (d) Schematic diagram of the interaction energy between different active sites and I₃⁻ in the N₄-COF. (e) Diagram of the differential charge at site 4 in the N₄-COF interacting with I₃⁻. (f) Difference in the DOS between I₃⁻ and I₂@N₄-COF. (g) TDOS plots of the N_x-COF.¹⁰⁹ Reproduced with permission from ref. 109. Copyright 2025, Wiley-VCH. (h) Schematic diagram of the synthesis route and molecular structure of COF-RuNCS-X. (i) Adsorption energy of I⁻, I₃⁻, and I₅⁻ for RuNCS. (j) Reaction mechanism of COF-RuNCS-X during the charge–discharge process.¹¹³ Reproduced with permission from ref. 113. Copyright 2025, Wiley-VCH.

COCs, as a new class of supramolecular materials, have demonstrated excellent adsorption properties due to their lightweight framework, accessible channels, outstanding stability, and tunable structure,¹¹⁴ making them particularly suitable for the selective capture of specific guest species. Compared to COFs and MOFs, COCs are simpler to synthesize, with superior structural stability and mechanical strength.¹¹⁵ Building on this, and combining the redox activity of pyridine compounds, Zhu et al.⁸³ reported a novel non-porous bipyridyl-based covalent organic cage (Bpd-COC) material, which was used as a cathode host material in aqueous Zn–I₂ batteries. This material was formed through supramolecular self-assembly of amine and bipyridyl monomers, resulting in a unique molecular cage structure. This cage framework not only maintains structural integrity but also enables effective confinement of I₃⁻ through interactions between the nitrogen atoms on the framework and polyiodide anions (Fig. 14a). DFT calculations show that Bpd-COC exhibits a lower adsorption energy for I₃⁻ than its precursor monomer (Fig. 14b), indicating that its molecular cage structure provides stronger confinement for iodine species. Unlike conventional porous activated carbon materials that primarily rely on physical adsorption, the Bpd-COC forms a more stable chemical adsorption interface through supramolecular interactions, effectively alleviating the shuttle effect of polyiodide anions during cycling and significantly suppressing corrosion of the zinc anode and the formation of by-products (Fig. 14c). Building upon this research, further studies not only continue to leverage the advantages of COCs in confining iodine species but also deepen the synergistic control mechanism between structural closure and nitrogen site density. Zhang et al.¹¹⁶ proposed the concept of “polyiodide reservoirs” by adjusting the nitrogen active site density around the three-dimensional (3D) cavity of the COC, creating a series of molecular confinement structures, including macrocycle M-1, cage C-2, and superphane S-3 (Fig. 14d), which evolve from an open to a nearly closed structure. Fourier-transform infrared spectroscopy (FTIR) and ¹H nuclear magnetic resonance (NMR) spectroscopy were used to characterize the chemical composition and molecular structure of these compounds, confirming their successful synthesis (Fig. 14e). Molecular dynamics simulations (Fig. 14f) and finite element analysis (Fig. 14g) revealed the confinement behavior of polyiodide anions in different molecular structures, which can be represented by the model shown in Fig. 14h. The results show that as the nitrogen active site density increases, iodine species gradually transition from a partially embedded state to being completely enclosed within the molecular framework. Particularly in the S-3 structure, the near-closed cavity and abundant nitrogen sites work synergistically to effectively confine iodine anions inside the molecule, significantly reducing the shuttle effect. Additionally, XPS (Fig. 14i) and in situ Raman results (Fig. 14j) revealed that the confinement structure stabilizes the redox pathway of iodine, following the reversible conversion process I⁻ ↔ I₃⁻ ↔ I₅⁻, thereby significantly enhancing the cycling stability and controllability of the reaction. Electrochemical testing further evaluated the rate performance of the structures macrocycle M-1, cage C-2, and superphane S-3, with superphane S-3@I₃⁻ exhibiting superior rate performance (Fig. 14k). Furthermore, superphane S-3, which readily dissolves in dimethyl sulfoxide (DMSO), can efficiently re-adsorb iodine species while maintaining good cycling performance and capacity retention under continuous cycling conditions. These synthesized materials combine structural confinement effects with dynamic recyclability, effectively suppressing iodine shuttle and enhancing electrode sustainability, offering new possibilities for the development of host materials in Zn–I₂ batteries.

Fig. 14 (a) Molecular and crystal structures of the Bpd-COC and the schematic diagram of I₃⁻ adsorption on the Bpd-COC. (b) Comparison of the adsorption energy of 2,2′-BPd-DA, Tren, and Bpd-COC for I₃⁻. (c) Diagram of the I₃⁻ shuttle suppression mechanism in a Bpd-COC-based aqueous zinc-ion battery (AZIB) compared to the conventional AC-based AZIB.⁸³ Reproduced with permission from ref. 83. Copyright 2025, Elsevier. (d) Fabrication process of COCs. (e) Digital images of M-1, C-2, and S-3 dissolved in H₂O and DMSO, along with their corresponding ¹H NMR spectra. Schematics of (f) adsorption process, (g) diffusion behavior, and (h) diffusion model of I₃⁻ in M-1@I₃⁻, C-2@I₃⁻, and S-3@I₃⁻. (i) I 3d XPS spectra of the S-3@I₃⁻ cathode at different voltages. (j) Time–voltage curves and the corresponding in situ Raman spectra of the S-3@I₃⁻ cathode during the charge and discharge process. (k) Rate performance of M-1@I₃⁻, C-2@I₃⁻, and S-3@I₃⁻ at various current densities.¹¹⁶ Reproduced with permission from ref. 116. Copyright 2025, Wiley-VCH.

Recently, PBAs have been used as iodine host materials in Zn–I₂ batteries due to their sufficient porosity and ordered microporous channels,¹¹⁷ where the strong chemical interaction between the small cross-linked pore size and I₂ can effectively promote electron/ion transport, assist in the utilization of I₂, and enhance the ability to confine iodine species in the pores, thus achieving high capacity output.⁸⁹ For example, Ma et al.¹¹⁸ developed a PBA material Co[Co_xFe_1−x(CN)₆] (0 ≤ x ≤ 1) with an ordered and continuous channel structure using a two-step method and incorporated Co and Fe dual-metallic electrocatalytic active sites (Fig. 15a). The Co and Fe atoms serve as synergistic electrocatalytic active centers that effectively block the path of I₂ to the intermediate product I₃⁻, enabling direct and efficient conversion of I₂ to I⁻, significantly inhibiting the formation and shuttle of polyiodide anions. SEM images (Fig. 15b) show that Co[Co_1/4Fe_3/4(CN)₆] has a regular truncated cubic morphology, indicating that the synthesized material has high crystallinity and structural order. Compared to Co[Co(CN)₆], Co[Fe(CN)₆], and non-catalytic porous carbon, Co[Co_1/4Fe_3/4(CN)₆] exhibited lower activation energy barriers and Gibbs free energy (Fig. 15c and d), indicating that I₂ undergoes the fastest kinetic conversion on Co[Co_1/4Fe_3/4(CN)₆] and exhibits stronger confinement of I₂. UV-vis absorption spectroscopy further confirmed (Fig. 15e) that the I₂ reduction reaction under Co[Co_1/4Fe_3/4(CN)₆] showed nearly no I₃⁻ formation, indicating that iodine species had been completely converted to I⁻, reflecting the superior confinement catalysis of this framework material. Meanwhile, its lowest Tafel slope (Fig. 15f) also validates the fastest electrochemical conversion rate of I₂/I⁻ in this host material from a kinetic perspective. The Zn–I₂ battery with Co[Co_1/4Fe_3/4(CN)₆] as the iodine host material demonstrated high capacity. This work showcases the confinement–catalysis–conduction coupling effect in regulating the iodine species conversion pathway, inhibiting polyiodide anion formation and accelerating charge transfer, providing a design approach for the development of host materials. Considering the role of pre-intercalated ions in cathodes in enhancing ion conductivity, improving material stability, and optimizing interfacial reactions, it may be beneficial to introduce pre-intercalated ions into the PB framework to modify its surface chemical environment, thereby regulating the redox behavior of halogen species. For example, Gao et al.⁴⁹ proposed an innovative structural confinement strategy by pre-confining iodine ions (PBI) into the PB framework to modify the surface chemical environment of PB materials, thereby effectively regulating the redox behavior of I⁻/I₃⁻, improving iodine conversion efficiency and utilization. XPS analysis shows (Fig. 15g and h) that PBI exhibits good reversibility of Fe²⁺/Fe³⁺ and I⁻/I₃⁻ during charge and discharge, with the I 3d spectra indicating stronger confinement of I₃⁻. UV-vis absorption spectra (Fig. 15i) further confirmed that there is almost no I₃⁻ signal in the PBI electrolyte, only I⁻, and the color change in the electrolyte also verified its effective suppression of polyiodide anions. DFT calculations (Fig. 15j) show that the ΔG for I⁻ oxidation in the PBI electrode is significantly lower than that in PB, further confirming its superior reaction kinetics. Moreover, the Tafel slope of the PBI//Zn full battery is only 155 mV dec⁻¹ (Fig. 15k), reflecting the fast kinetics of the I⁻/I₃⁻ redox reaction. The above research fully demonstrates that by rationally designing metal site synergistic effects and structural pre-embedding mechanisms, not only can the I₂/I⁻ and I⁻/I₃⁻ reaction pathways be effectively controlled, but iodine utilization efficiency and the electrochemical performance of the battery can also be significantly improved, providing new ideas for building high-performance, long-life aqueous Zn–I₂ batteries.

Fig. 15 (a) Diagram of the synthesis of PBAs/I₂ and the electrochemical process involving catalytically active Co and Fe sites. (b) SEM image of Co[Co_1/4Fe_3/4(CN)₆]. (c) Activation energy comparison chart of porous carbon, Co[Co(CN)₆], Co[Fe(CN)₆], and Co[Co_1/4Fe_3/4(CN)₆], at the onset potential of the halogen reduction reaction. (d) Comparison of Gibbs free energy for the I₂ reduction reaction on Co[Co(CN)₆], Co[Fe(CN)₆], Co[Co_1/4Fe_3/4(CN)₆] and graphene. (e) UV-Vis absorption spectra of electrolytes corresponding to porous carbon, Co [Fe(CN)₆], Co[Co(CN)₆] and Co[Co_1/4Fe_3/4(CN)₆] as host materials after sufficient IRR. (f) Corresponding tafel plots derived from cyclic voltammetry (CV) curves of porous carbon, Co[Co(CN)₆], Co[Fe(CN)₆], and Co[Co_1/4Fe_3/4(CN)₆].¹¹⁸ Reproduced with permission from ref. 118. Copyright 2020, Wiley-VCH. Core-level spectra of (g) Fe 2p and (h) I 3d for PBI and PB electrodes at various charge and discharge states. (i) UV-vis spectra of PB/PISS (PISS: 1 M Na₂SO₄ + 0.1 M H₂SO₄ + 25 × 10⁻⁴ M KI) and PBI/PISS after the cycling test. (j) Comparative Gibbs free energy of the I⁻ oxidation process in PBI and PB. (k) Tafel plots of Zn-based batteries assembled with PBI and PB cathodes.⁴⁹ Reproduced with permission from ref. 49. Copyright 2023, Wiley-VCH.

In addition to the typical functional frameworks mentioned above, porous aromatic frameworks (PAFs) and covalent triazine frameworks (CTFs) have also been explored as iodine host materials in Zn–I₂ batteries. PAFs are constructed using aromatic units, forming a high specific surface area and uniform pore size distribution, while also possessing high rigidity and excellent chemical resistance.¹¹⁹ In the PAF structure, abundant benzene rings provide affinity for polyiodides and promote electron and ion transfer.⁵³ To this end, Hu et al.⁵³ introduced biphenyl struts through the Yamamoto coupling reaction to replace C–C bonds, constructing an aromatic porous framework material (PAF-1) with a rich benzene ring structure and ordered channels (Fig. 16a). Benefiting from its high surface area and interconnected 3D channel structure, this material demonstrates significant advantages in confining polyiodide anions. EPR results (Fig. 16b) show the presence of free radicals in I₂@PAF-1, confirming a strong charge transfer interaction between PAF-1 and polyiodide anions. Further X-ray absorption spectroscopy (XAS) analysis (Fig. 16c and d) reveals that iodine species in I₂@PAF-1 exhibit a mixed valence state ranging from −1 to 0, with iodine–iodine bond lengths longer than in pure iodine, indicating the stable presence of polyiodide anions in the framework, showcasing its excellent confinement capability. Electrochemical testing of the battery assembled with this material as the cathode host (Fig. 16e) shows typical I⁻/I₂ two-electron conversion behavior in CV curves (Fig. 16f) and good rate performance at various current densities (Fig. 16g). Galvanostatic intermittent titration technique (GITT) tests further verify that iodine species within the PAF-1 framework exhibit fast diffusion kinetics (Fig. 16h), significantly outperforming traditional “rocking-chair” battery systems. Moreover, comparing PAF-1 with other PAF materials with different pore structures highlights the synergistic advantages of the abundant benzene rings as anchoring sites and the high surface area in the 3D channels. This work systematically validates the unique structural advantages of PAF-1 in confining polyiodide ions, demonstrating the vast application potential of aromatic porous framework materials in Zn–I₂ energy storage systems and providing new directions for material selection and design for high-energy-density, long-cycle-life aqueous Zn–I₂ batteries. Furthermore, CTFs, characterized by triazine units, also show great potential for iodine adsorption due to their rich triazine units, highly uniform porosity, and excellent physicochemical stability.¹²⁰ However, the interaction between CTFs and iodine species is relatively weak, and they cannot effectively suppress the dissolution and shuttle of iodine species (especially I₃⁻ and I₅⁻) during electrochemical reactions.¹²¹ Based on this, Zhao et al.¹²² designed a pyridine-rich cationic covalent triazine framework (CCTF-TPMB) as an iodine host material. This material is not only scalable but also possesses a highly uniform pore structure and large surface area, favoring rapid mass transport. In situ UV-vis spectroscopy results (Fig. 16i and j) show that the I₃⁻ signal in the electrolyte is extremely weak during continuous charge and discharge of I₂@CCTF-TPMB, indicating that the abundant pyridine cationic sites in the material effectively confine iodine species and suppress their shuttle. Theoretical calculations (Fig. 16k) confirm that during the iodine reduction process, ΔG for the CCTF-TPMB system is more negative, indicating that the reduction from I₂ to I⁻ is more spontaneously favorable and the polyiodide conversion rate is higher. Additionally, CCTF-TPMB shows strong adsorption ability for various iodine species (Fig. 16l), effectively promoting the stable confinement of iodine species. Electronic state analysis (Fig. 16m–p) reveals that iodine species on CCTF-TPMB exhibit a wider bandgap distribution and a shift towards lower energy states, reflecting more stable electronic states and adsorption states, further indicating its excellent confinement effect and electrochemical activity at the electronic structure level. These comprehensive results demonstrate that the pyridine-rich CCTF-TPMB framework not only significantly suppresses iodine species shuttle but also promotes rapid and reversible electrochemical conversion. This provides valuable guidance for exploring the structure–activity relationship of cathode catalysts with well-defined active sites and offers important material design insights for the development of low-cost, high-performance Zn–I₂ batteries.

Fig. 16 (a) Schematic diagram of the diamond-like topology of PAF-1. (b) EPR spectra of I₂@PAF-1, PAF-1 and I₂. (c) X-ray absorption near edge structure (XANES) spectra of I₂@PAF-1, KI, CsI₃, and I₂. (d) Normalized extended X-ray absorption fine structure (EXAFS) spectra of I₂@PAF-1, KI, CsI₃, and I₂. (e) Schematic diagram of the structure and reaction mechanism of the Zn–I₂ battery with I₂@PAF-1 as a cathode. (f) CV curve of the Zn–I₂@PAF-1 battery at a scan rate of 0.1 mV s⁻¹. (g) Rate performance and CE of the Zn–I₂@PAF-1 battery at 0.5–8 C. (h) I⁻ diffusion coefficients and GITT profiles of the I₂@PAF-1 cathode at 0.5 C.⁵³ Reproduced with permission from ref. 53. Copyright 2024, Wiley-VCH. In situ UV-vis spectra of the electrolytes corresponding to (i) I₂@CCTF-TPMB and (j) I₂@CTF-DCBP cathodes during the charge–discharge process. (k) Gibbs free energy of the IRR on CTF-DCBP and CCTF-TPMB. (l) Adsorption energy of I⁻, I₂, I₃⁻, and I₅⁻ on CTF-DCBP and CCTF-TPMB. Comparison of the PDOS of the I 5p orbital for species (m) I⁻, (n) I₂, (o) I₃⁻, and (p) I₅⁻ before and after adsorption on CTF-DCBP and CCTF-TPMB.¹²² Reproduced with permission from ref. 122. Copyright 2024, Wiley-VCH.

In summary, functional framework materials such as MOFs, COFs, COCs, PBAs, PAFs, and CTFs exhibit exceptional iodine species confinement and conversion regulation capabilities in Zn–I₂ batteries due to their high surface area, tunable pore structures, abundant chemically active sites, and excellent structural stability. Additionally, certain host materials enhance the electron transfer process of iodine species by constructing conductive channels or introducing synergistic catalytic active centers, thereby achieving high energy density and long cycle life Zn–I₂ batteries. Despite this, future efforts should focus on the following key scientific issues: (i) further analyzing the relationship between structural configurations and iodine confinement behavior to better understand the effects of pore size distribution, surface polarity, and their influence on the adsorption and conversion pathways of iodine species; (ii) developing multi-scale in situ characterization techniques (such as synchrotron XAS, in situ Raman/IR, electron tomography, etc.) to monitor the distribution, conversion, and dynamic evolution of iodine species within the framework channels in real-time; (iii) combining first-principles calculations and machine learning (ML) tools to achieve predictions and optimization designs of functional sites, accelerating the screening and development of high-performance framework host materials. With the continuous optimization of structural regulation and the integration of cross-scale research methods, functional framework materials are expected to provide a new paradigm for the development of aqueous Zn–I₂ batteries.

3.1.3. Derivatives of functional frameworks. In addition to the typical functional frameworks, carbon-based materials derived from functional frameworks have garnered attention due to their tunable structures and diverse functionalities, making them highly attractive for the confinement and redox conversion of iodine species.³³ Among these, MOFs, constructed from metal ions or clusters and organic ligands, serve as ideal precursors for preparing highly refined derived carbon materials. During pyrolysis, decarboxylation reactions and metal redox processes within MOFs facilitate the formation of hierarchical pore structures, giving the derived carbon materials unique morphologies and rich porosity.¹²³ More importantly, MOF-derived carbon retains some of the structural advantages of the original framework. By precisely controlling the composition of the MOF precursor, metal center types, and organic ligand configurations, it is possible to fine-tune the conductivity, specific surface area, and functional group distribution of the resulting carbon material,¹²⁴ optimizing its electrochemical performance and providing strategic support for constructing high-performance iodine host materials. For instance, Chai et al.¹²⁵ used indium-based MOFs (InOF-1) as precursors and successfully synthesized hollow carbon nanotubes (HCNS) with a porous structure through pyrolysis in an inert atmosphere, driven by reduction between the carbon matrix and nanoscale indium oxide (Fig. 17a). During the carbonization process, the melting and migration of indium nanoparticles and decarboxylation of organic ligands occur simultaneously. The metal components are effectively removed at high temperatures, leading to the in situ formation of HCNS with hollow structures and interconnected pores. While CNTs can load small iodine species and buffer volume changes, their dense walls limit the ability to capture small molecules, and iodine species within CNTs can only migrate along limited paths, reducing ion transport efficiency and material utilization. In contrast, HCNS not only has a hollow structure but also retains the microporous structure of the MOF precursor and the meso/macroporous structure formed, providing more active sites and shortening the diffusion paths of active materials and ions. XPS analysis reveals the chemical structure changes of InOF-1 during heat treatment. The C 1s spectrum (Fig. 17b) shows peaks for C=C, C–C, and –COO, indicating that the organic ligand structure is well retained. As the temperature increases, the –COO peak weakens, and a C–O/C–N peak appears at 286.4 eV, signaling decarboxylation, which provides a porous carbon structure for subsequent iodine confinement. The In 3d spectrum (Fig. 17c) shows that In₂O₃ undergoes reduction during heat treatment, generating metallic indium, which volatilizes and removes during the subsequent process, confirming the formation mechanism of the hollow structure. Electrochemical impedance spectroscopy (EIS) results (Fig. 17d) show that HCNS exhibits lower charge transfer impedance, series resistance, and higher Warburg impedance slope compared to CNTs, indicating superior conductivity and ion diffusion. These properties together demonstrate the significant advantages of HCNS as a cathode host material for iodine species. Its high surface area and internal hierarchical pore structure enable effective confinement and stable loading of iodine species, significantly enhancing electrode reactivity and cycling stability. This work demonstrates that MOF-derived hollow porous carbon nanomaterials with excellent confinement effects can be synthesized, opening new pathways for developing highly ordered and multifunctional framework-derived carbon materials. Notably, the confinement performance of MOF-derived carbon materials depends not only on the hierarchical regulation of their pore structures and hollow morphology but also on the precursor size control and thermal treatment conditions,¹²⁶ which play a critical role in adjusting iodine species confinement and storage behavior. By adjusting the scale of the MOF structural units and carbonization conditions, pore size distribution, specific surface area, and graphitization degree can be further optimized, achieving efficient confinement and fast conversion of iodine species.¹²⁷ Based on this, Li et al.¹²⁸ synthesized a series of Zn-MOF-74 rod-like precursors, labeled P1–P5, via solvothermal reaction, and then conducted high-temperature carbonization to obtain porous carbon nanorods (PCN) derived from Zn-MOF-74. By controlling the carbonization temperature, they obtained samples P2-900, P2-1000, and P2-1100 and used a size-controllable strategy to confine iodine species in the pores (I₂@PCN), reducing the formation of I₃⁻ and I₅⁻ and accelerating the conversion of polyiodide species (Fig. 17e). Raman spectroscopy (Fig. 17f) shows that as the carbonization temperature increases, the graphitization degree of PCN enhances, improving its electronic conductivity. Nitrogen adsorption–desorption tests (Fig. 17g and h) further confirm that after pyrolysis, the material possesses abundant micropores and mesopores with a high specific surface area, capable of accommodating I⁻, I₂, and I₃⁻ iodine species. Among them, P2-1000 has the highest specific surface area and favorable pore size distribution, which is advantageous for iodine species loading and confinement. The rate performance of the Zn–I₂ battery with I₂@PCN as the cathode is shown in Fig. 17i, where I₂@P2-1000 exhibits the best rate performance. In situ UV-vis (Fig. 17j) and Raman spectroscopy (Fig. 17k) techniques further reveal that I₂@P2-1000 not only effectively suppresses the generation and diffusion of soluble species like I₃⁻ and I₅⁻ but also accelerates their conversion to I⁻ during the discharge process, indicating that its suitable pore structure and surface chemistry play a key role in stabilizing iodine species. The soft-pack battery based on the I₂@P2-1000 cathode maintained a specific capacity of 141.7 mAh g⁻¹ after 100 cycles at a current density of 100 mA g⁻¹ (Fig. 17l). Additionally, the assembled soft-pack battery with the I₂@P2-1000 cathode can power devices, showing strong potential for applications in flexible electronics. This study demonstrates that adjusting the size of MOF precursors and the carbonization temperature to optimize pore size distribution and maximize specific surface area enables porous carbon frameworks to significantly enhance iodine confinement and improve reaction kinetics. These findings confirm the feasibility of MOF-derived carbon materials for enhancing redox reaction kinetic characteristics and effectively confining polyiodide shuttling, providing versatile strategies for designing high-performance, high-conductivity, and stable carbon-based iodine host materials.

Fig. 17 (a) Energy storage mechanism of CNTs and HCNS during the charge–discharge process. XPS spectra of (b) C 1s and (c) In 3d of InOF-1@T. (d) Nyquist plots of HCNS and CNT.¹²⁵ Reproduced with permission from ref. 125. Copyright 2022, Wiley-VCH. (e) Fabrication process of I₂@PCN. (f) Raman spectra of P2-1100, P2-1000, and P2-900. (g) Pore characteristics of 1: P1-1000, 2: P2-1000, 3: P3-1000, 4: P4-1000 and 5: P5-1000. (h) N₂ adsorption–desorption curves of P2-1100, P2-1000, and P2-900. (i) Rate performance of the I₂@P2-1000 cathode under different current densities. (j) GCD curve of I₂@P2-1000 in ZnSO₄ solution and the corresponding in situ UV-vis spectrum. (k) In situ Raman spectra and associated contour maps of I₂@P2-1000. (l) GCD curves of the Zn/I₂@P2-1000 pouch cell.¹²⁸ Reproduced with permission from ref. 128. Copyright 2024, Wiley-VCH.

To enhance the confinement and catalytic conversion of iodine species in functional framework-derived carbon materials, researchers have gradually incorporated SACs with higher catalytic activity. By dispersing single-atom metals onto porous carbon substrates derived from framework materials, composite materials with both spatial confinement and electrocatalytic functions can be constructed,¹²⁹ which are expected to further suppress polyiodide shuttling and achieve efficient catalytic conversion of iodine species and key intermediates. For example, Liu et al.³¹ reported a novel non-porous bipyridine-based covalent organic cage (Bpd-COC) material derived from melamine-cyanuric acid complexes, synthesized by pyrolysis to form a metal–nitrogen–carbon (Fe–N–C) atomic bridge structure. This structure not only possesses abundant micro/mesoporous channels and a large surface area, facilitating efficient iodine adsorption and rapid diffusion, but also introduces electron-accepting nitrogen atoms to modulate the surface electronic density of neighboring carbon atoms, significantly enhancing the polarity of iodine species adsorption and overcoming the limitations of traditional carbon materials, which mainly rely on physical adsorption for iodine confinement. In the material, Fe atoms are dispersed in the form of Fe–N₄ coordination, and their valence state ranges from 0 to +3. XPS and EXAFS analyses did not detect Fe–Fe coordination, confirming the presence of single-atom structures (Fig. 18a and b). DFT calculations showed that Fe–N₄–C has the lowest adsorption energy for I₂ (−1.18 eV), demonstrating the most stable adsorption behavior (Fig. 18c and d). Charge density difference analysis (Fig. 18e and f) revealed significant electron rearrangement between Fe–N₄ centers and I₂. PDOS and band center calculations (Fig. 18g and h) revealed that the orbital hybridization effect between Fe–N₄–C and I₂ is the strongest, facilitating charge transfer and improving reaction kinetics. Moreover, its Gibbs free energy for the I₃⁻ → I⁻ conversion process is the lowest (ΔG = 0.89 eV), significantly lower than that of N–C (0.92 eV) and C (0.99 eV) (Fig. 18i), highlighting its unique advantage in suppressing polyiodide shuttling and enhancing redox reversibility (Fig. 18j). In battery performance testing, the Zn–I₂ battery using B–Fe–NC as the cathode exhibited a high capacity of 172 mAh g⁻¹, over 5000 cycles of lifespan, and nearly 99.7% CE, significantly outperforming similar systems reported at the time. Similarly, Guo et al.³³ designed a series of iron-doped Fe–N–C materials (M9) derived from Zn-MOF precursors and fabricated through a simple self-assembly-carbonization strategy, used as iodine host materials in Zn–I₂ batteries (Fig. 18k). In M9, Fe atoms form stable coordination with nitrogen atoms, and the pore structure of ZIF-8 provides molecular-level confinement space, effectively encapsulating metal precursors and ensuring atomic-level dispersion. Fourier transform EXAFS and wavelet transform results show that Fe atoms in M9 exist in the form of Fe–N coordination, rather than Fe–Fe coordination, indicating the successful formation of single-atom Fe–N₄ active centers (Fig. 18l). This confinement structure not only enhances the electronic coupling between Fe and iodine species but also significantly improves the electrocatalytic activity of the material. EXAFS fitting further confirmed the coordination environment of Fe–N₄ and showed excellent fitting precision. Theoretical calculations demonstrated that M9 has a lower energy barrier in the iodine reduction reaction, indicating that its adsorption and catalytic conversion of I⁻ is significantly better than that of the control samples (Fig. 18m). Spin density analysis revealed a significant accumulation of unpaired electrons around the Fe atoms, indicating their higher electrical conductivity (Fig. 18n), which accelerates the redox reaction kinetics of iodine species and demonstrates the high confinement-catalysis performance provided by Fe–N–C. The studies highlight the advantages of these functional framework-derived materials in the confinement and stabilization of iodine intermediates, while significantly lowering the reaction energy barrier. This dual synergistic control of confinement and electrocatalysis provides important design insights for developing advanced cathode materials for high-performance Zn–I₂ batteries.

Fig. 18 (a) N 1s XPS spectra of B–Fe–NC. (b) Fourier transformation of EXAFS spectra in R-space for B–Fe–NC and reference materials including Fe foil, FePc, and Fe₂O₃. Charge density distribution of Fe–N₄–C (c), N–C, and pure carbon (d) surfaces, along with the adsorption energy of I₂ on these structures. Mulliken charge distributions on Fe–N₄–C before (e) and after (f) I₂ adsorption. PDOS of the iodine p-band in C–I₂ and NC–I₂ (g) and Fe d-band in B–Fe–NC–I₂ (h), respectively. (i) Comparative Gibbs free energy for the I₂ reduction reaction occurring on Fe–N₄–C, N–C, and C substrates. (j) Reaction mechanism of polyiodide adsorption/conversion in the B–Fe–NC host material.³¹ Reproduced with permission from ref. 31. Copyright 2022, American Chemical Society. (k) Fabrication process of M–N–C. (l) Fourier-transform EXAFS spectra of M9 alongside Fe foil and FePc reference samples. (m) Diagrams of Gibbs free energy changes during I₂ reduction reaction on M4 (N–C) and M9. (n) Spin density analysis and surface ESP visuals for M4 and M9.³³ Reproduced with permission from ref. 33. Copyright 2024, Wiley-VCH.

To understand the role of functional framework-derived carbon confinement structures and atomic-level catalytic sites in the synergistic confinement and efficient catalytic conversion of iodine species, Yang et al.¹³⁰ used mesoporous UiO-66-NH₂ as a precursor and pyrolyzed it to obtain iodine host materials (Fe SAC-MNC) with well-dispersed single-atom Fe sites (Fe SAC) in ordered mesoporous nitrogen-doped carbon (MNC). This material efficiently confines and converts I₂/I⁻ and polyiodide anion intermediates (I₃⁻). Traditional disordered activated carbon (AC) shows some conductivity but suffers from low iodine loading, slow reaction kinetics, and insufficient ability to suppress I₃⁻ shuttling (Fig. 19ai). Ordered mesoporous carbon frameworks offer some advantages in increasing iodine loading and mitigating intermediate diffusion; however, due to the lack of interface catalytic activity, their inherent redox rates remain limited (Fig. 19aii). The Fe SAC-MNC, however, employs a dual mechanism of pore confinement and atomic-level catalysis, significantly lowering the activation energy barrier of the I₂ electro-reduction reaction and accelerating the kinetics of multi-step electrochemical processes involving I₂/I₃⁻/I⁻ (Fig. 19aiii). Microstructural characterization confirmed that Fe SAC sites are highly uniformly distributed within the carbon framework (Fig. 19b). DFT calculations indicated that this structure exhibits the strongest physical–chemical adsorption for I⁻, I₂, and I₃⁻, significantly enhancing the confinement of intermediates and effectively inhibiting shuttling behavior (Fig. 19c). UV-vis absorption spectroscopy further confirmed that the Fe SAC-MNC shows a better I₂ binding affinity compared to MNC and KB control materials (Fig. 19d). Kinetic analysis revealed that the Fe SAC-MNC has a lower IRR activation energy and almost no I₃⁻ intermediate formation during the reaction, allowing for the direct conversion of I₂ to I⁻, preventing diffusion losses of active materials due to intermediates (Fig. 19e). Fig. 19f compares the activation energy of different catalysts for the I⁻/I₂ conversion reaction, showing that the Fe SAC-MNC exhibits the lowest activation energy (E_a = 27.878 kJ mol⁻¹), indicating its superior catalytic kinetics in promoting iodine redox reactions. This demonstrates the potential of single-atom confinement catalysis in improving electrochemical conversion efficiency. Free energy calculations further show that the ΔG of key conversion steps in this system is significantly lower than that of other hosts, making the reaction more favorable and accelerating its kinetics (Fig. 19g). Electrochemical performance tests also validate the proposed mechanism: Fe SAC-MNC electrodes exhibit excellent specific capacity and rate performance at various current densities, significantly outperforming comparison electrodes and maintaining good stability even at high rates (Fig. 19h). Furthermore, its cycling performance remains highly stable, retaining a high capacity and close to 100% CE after long-term operation (Fig. 19i), demonstrating the comprehensive advantages of Fe SAC sites in enhancing electron transfer efficiency, iodine species conversion efficiency, and capacity retention. This work showcases the critical role of atomic-level dispersed metal sites and mesoporous carbon frameworks in regulating the I₂/I⁻ redox pathway, providing a valuable design strategy for constructing Zn–I₂ batteries with high energy density and good cycling stability.

Fig. 19 (a) (ai) Redox reaction mechanism in aqueous Zn||I₂ batteries, (aii) “confinement-catalysis” design concept of the iodine cathode, (aiii) illustration of decreased conversion energy barrier via an embedded Fe SAC. (b) HAADF-STEM image of Fe SAC-MNC. (c) Adsorption energy of I⁻, I₂, and I₃⁻ on Fe SAC-MNC, MNC, and KB, and the corresponding charge density difference maps of I⁻, I₂, and I₃⁻ on an Fe SAC-MNC. (d) UV-vis absorption spectra of I₂ in Fe SAC-MNC/I₂, MNC/I₂, and KB/I₂ systems with 2 M ZnSO₄. (e) Activation energy comparison for IRR among Fe SAC-MNC, MNC, and KB. (f) UV-vis absorption spectra of electrolytes of KB, MNC, and Fe SAC-MNC after complete IRR. (g) Gibbs free energy profiles for the I₂ reduction reaction on Fe SAC-MNC, MNC, and KB. (h) Rate performance of Fe SAC-MNC/I₂, MNC/I₂, and KB/I₂ cathodes under different current densities. (i) Long-term cycling stability of Zn||KB/I₂, Zn||MNC/I₂, and Zn||Fe SAC-MNC/I₂ batteries at a current density of 1 A g⁻¹.¹³⁰ Reproduced with permission from ref. 130. Copyright 2023, Springer Nature.

Cobalt possesses a unique d⁷ electronic configuration and flexible oxidation states, enabling rapid electron transfer and accelerating catalytic reactions in redox processes.¹³¹ Recently, Han et al.¹³² developed a dual-function halogen host material, Co/C800, derived from ZIF-67 as the precursor, which was pyrolyzed to create a cobalt-doped nitrogen-doped carbon composite (Fig. 20a). The high specific surface area and porous characteristics of ZIF-67-derived carbon provide abundant active sites for both adsorption and catalytic reactions. EIS tests indicate that when I₂@Co/C800(HCl) is used as the cathode, its charge transfer resistance is significantly lower than that of undoped samples, suggesting that the incorporation of cobalt helps accelerate electrode reaction dynamics (Fig. 20b). GCD tests show that, under the same current density, the electrode exhibits a higher average discharge capacity, outperforming other comparative samples (Fig. 20c). Furthermore, the presence of cobalt effectively lowers the charging voltage and enhances the discharge plateau, further demonstrating its ability to regulate the electrochemical reaction pathway (Fig. 20d). To further investigate the redox mechanisms of halogen species, the researchers monitored reaction intermediates during charge and discharge cycles using in situ Raman and UV-vis spectroscopy. The results showed that Raman peaks corresponding to I₃⁻ and I₅⁻ appeared during charging and gradually disappeared during discharging, indicating the reversible conversion of iodine species within the host material (Fig. 20e). XPS analysis revealed the chemical state evolution of cobalt and nitrogen elements at different voltage states: initially, Co 2p_3/2 displayed peaks corresponding to metallic and low-valent states; during charging, the high-valent peaks increased, indicating that high-valent cobalt aids in adsorbing charged intermediates; during discharge, the peaks returned to their initial state, reflecting excellent reversibility (Fig. 20f and g). Simultaneously, nitrogen also exhibited reversible valence changes during charging and discharging, facilitating the co-adsorption of halogen species (Fig. 20h). This cobalt/nitrogen co-doped carbon-based material not only enhanced the confinement ability for polyhalide intermediates by tuning the electronic structure but also significantly accelerated their redox reaction kinetics, exhibiting good confinement–catalysis–conduction properties of I₂@Co/C800(HCl) (Fig. 20i and j). Under this mechanism, the assembled Zn–I₂ battery demonstrated high specific capacity and excellent cycling stability, maintaining good capacity retention after long-term cycling. This study validates the synergistic effect of co-doped framework-derived carbon materials in regulating interface electronic structures and enhancing iodine species fixation and conversion performance, guiding the rational design of high-efficiency, reversible iodine cathode host materials.

Fig. 20 (a) Synthesis process of I₂@Co/C800 removes Co. (b) EIS curves of I₂@Co/C800(HCl), I₂@Co/C800 and I₂@Co/C800(HCl–HNO₃). (c) Cycling performance and CE of I₂@Co/C800(HCl), I₂@Co/C800 and I₂@Co/C800(HCl–HNO₃) at 5C. (d) GCD curves of I₂@Co/C800(HCl), I₂@Co/C800 and I₂@Co/C800(HCl–HNO₃). (e) In situ Raman spectra of the I₂@Co/C800(HCl) cathode during the charge–discharge process. XPS spectra of the I₂@Co/C800(HCl) cathode at different charge and discharge states, including (f) Co 2p, (g) I 3d, and (h) N 1s. Mechanism of batteries with Co/C800(HCl) (i) and Co/C800(HCl–HNO₃) (j) as cathodes, respectively.¹³² Reproduced with permission from ref. 132. Copyright 2024, Elsevier.

In summary, carbon-based composites derived from functional framework materials offer significant advantages in the design of iodine-based cathodes for Zn–I₂ batteries. By using ordered, composition-tunable precursors such as MOFs and ZIFs, the pyrolyzed materials exhibit hierarchical pore structures that enhance specific surface area and electronic/ionic transport efficiency while also achieving spatial confinement and structural stability of active iodine species. Notably, current research has evolved from single-dimensional design to a new paradigm of constructing multi-dimensional confinement structures combined with atomic-level electronic structure regulation for synergistic control. Further research can focus on the following specific aspects: (i) fine construction of multi-interface heterogeneous structures at the confinement scale, precisely regulating the spatial distribution and reaction paths of iodine species through interface charge modulation and local electric field guidance; (ii) construction of multifunctional active sites, such as dual metal sites, single-atom clusters, and defect engineering, to explore coupled-enhanced electronic structures and catalytic mechanisms; (iii) different characterization techniques, such as advanced in situ characterization methods, cryo-electron microscopy (cryo-TEM), and first-principles calculations, to deeply analyze the adsorption behavior, charge transfer pathways, and evolution mechanisms of key iodine intermediates in confinement structures; (iv) combining AI-assisted high-throughput screening and structural prediction of iodine-confining host materials to enhance material design efficiency and accelerate the engineering and practical applications of high-efficiency, stable Zn–I₂ battery systems.

3.1.4. 2D transition metal carbides/nitrides. In Zn–I₂ batteries, the low binding strength of iodine species to conductive supports leads to reduced electron transfer efficiency and severe shuttle effects.³⁸ To address these issues, Li et al.¹³³ proposed a method to capture I₂ within the nanoscale interlayer of the Nb₂CT_x MXene using electrodeposition, achieving in situ loading and confinement of active materials (Fig. 21a). The GCD curve and UV-vis results (Fig. 21b and c) show that no characteristic absorption peaks of I₃⁻ were observed throughout the reaction process, indicating direct reversible conversion between I⁻ and I₂, effectively avoiding the generation of the intermediate product I₃⁻. This confirms that Nb₂CT_x successfully confines and stabilizes iodine species, significantly reducing the loss of active materials. XRD results further show that the (002) characteristic diffraction peak of Nb₂CT_x remains present at different charge/discharge states, with the interlayer spacing increasing during discharge and returning to its original state upon charging, verifying its good structural reversibility and stability (Fig. 21d). Moreover, the Nb, C, and O (T_x) sites on the Nb₂CT_x surface show negative adsorption energies for I⁻, I₂, and I₃⁻, indicating a highly spontaneous adsorption process and confirming the inherent adsorption affinity of Nb₂CT_x's layered structure for iodine species (Fig. 21e). DOS results further confirm stable host–guest interactions between the Nb₂CT_x material and iodine species (Fig. 21f). These findings indicate that Nb₂CT_x serves as a stable and efficient iodine confinement host, combining excellent conductivity with structural reversibility, maintaining 80% of its capacity after over 23 000 cycles, demonstrating exceptional cycling stability. However, it is noteworthy that conventional multi-layer MXenes, due to their self-stacking effect, have relatively limited surface area utilization, restricting their ability to confine iodine species. In this context, Li et al.⁸⁶ developed a Ti₃C₂T_x (where T_x represents surface functional groups such as –O and –F) MXene-assisted high iodine-loaded composite cathode (MX-AB@I) (Fig. 21g). This composite consists of a mixed network of BC nanofibers and MXene nanosheets uniformly coated onto the surface of AC microparticles, forming continuous electronic transport channels between AC particles, thereby effectively enhancing the conductivity of the cathode (Fig. 21h). Theoretical calculations further show that the Ti₃C₂T_x MXene substrate exhibits significantly higher adsorption energies for I⁻, I₂, and I₃⁻ compared to the activated carbon substrate, especially showing superior iodine species confinement capability in suppressing I₃⁻, reflecting its strong adsorption and fixation of soluble polyiodide anions (Fig. 21i). After multiple cycles, compared to the control electrode without the MXene (AB@I), the MX-AB@I electrode significantly suppressed the shuttle effect of polyiodide anions and the zinc dendrite formation and irreversible side reactions induced by them, thereby improving the reversible deposition/stripping behavior of the Zn anode and enhancing the cycling stability of the battery (Fig. 21j and k). Thus, when used as the iodine host material for Zn–I₂ batteries, this material achieves higher zinc discharge depth and energy density, surpassing most of the Zn–I₂ battery systems reported in the literature. These studies demonstrate that MXenes have significant advantages in enhancing electronic conductivity and confining polyiodide anions, effectively alleviating the I₃⁻ shuttle effect and the instability of the Zn anode. However, under high-concentration electrolytes or long-cycle conditions, challenges remain in effectively suppressing side product migration and dendrite growth, as well as other uncontrollable issues.¹³⁴

Fig. 21 (a) Flowchart of the preparation of an electrodeposited I₂–Nb₂CT_x MXene electrode. (b) GCD curve of an I₂–Nb₂CT_x MXene cathode with marked voltage states, (c) UV-vis absorption spectra of the electrolyte at the corresponding voltage states after 10 000 cycles, and (d) XRD patterns of the electrode material. (e) Schematic of the Nb₂CT_x crystal structure and adsorption energy of I⁻, I₂, and I₃⁻ at different sites on Nb₂CT_x. (f) DOS of Nb₂CT_x before and after adsorption of I⁻, I₂, and I₃⁻ species at Nb sites.¹³³ Reproduced with permission from ref. 133. Copyright 2021, Wiley-VCH. (g) Fabrication process of the MX-AB@I cathode and the advantages as an iodine host material. (h) SEM image of MX-AB. (i) Comparison of adsorption energy for I⁻, I₂, and I₃⁻ on Ti₃C₂T_x and AC. Mechanistic diagrams of Zn–I₂ batteries with the (j) AB@I cathode and (k) MX-AB@I cathode.⁸⁶ Reproduced with permission from ref. 86. Copyright 2024, Wiley-VCH.

To address the aforementioned challenges, Yan et al.¹³⁵ proposed a ternary synergistic optimization strategy that combines the Ti₃C₂T_x MXene (TMX) as the iodine cathode host, n-butanol as an electrolyte additive, and a synergistic regulation mechanism of the in situ formed SEI layer. The synthesized TMX exhibits an ultra-thin layered structure with abundant surface active sites, contributing to the effective confinement of polyiodide species (Fig. 22a and b). UV-vis absorption results indicate that TMX shows a strong adsorption ability towards I₃⁻, significantly inhibiting its migration into the electrolyte (Fig. 22c). DFT calculations further confirm that TMX has a much lower adsorption energy for I₃⁻ compared to carbon cloth (CC) (Fig. 22d), demonstrating its strong thermodynamic affinity for polyiodide species. Notably, the interaction between TMX and I₃⁻ leads to significant electron enrichment, forming a strong host–guest interaction that effectively restricts I₃⁻ migration and reduces side reactions (Fig. 22e and f). Additionally, the partially dissolved n-butanol additive further collaborates with TMX, utilizing its strong adsorption capacity towards I₃⁻ to limit its migration (Fig. 22g). On the other hand, n-butanol and I⁻ in the electrolyte synergistically improve the solvation structure of Zn²⁺ (Fig. 22h). Under the influence of these multiple mechanisms, the constructed Zn–I₂ battery maintains high specific capacity and energy density after long cycling, demonstrating excellent cycling stability and good self-discharge suppression ability. Given the outstanding performance of MXene materials in inhibiting iodine species shuttle effects and achieving efficient confinement, researchers have gradually expanded their focus to other emerging 2D material systems. By regulating key structural parameters such as electronic structure, surface functional groups, and crystal defects, they aim to further enhance their chemical adsorption capacity and electrocatalytic conversion efficiency for iodine species,¹³⁶ thereby improving the performance of Zn–I₂ batteries. Recently, Zhang et al.¹³⁷ introduced Mo_4/3B₂T₂ (T representing O, –OH, and –F end groups) MBene materials with an in-plane ordered metal vacancy structure as iodine species host materials (Fig. 22i). The electron density difference map (Fig. 22j) reveals significant charge transfer between Mo atoms and iodine upon adsorption of iodine species on the Mo_4/3B₂T₂ surface, indicating strong electronic coupling ability. Further adsorption energy calculations show that Mo_4/3B₂T₂ exhibits a significantly better adsorption energy for I₂, I₃⁻, and other typical iodine species compared to Mo₂B₂T₂ and other common iodine host materials (such as graphene, MOF, etc.), demonstrating its stronger affinity and stable confinement ability for iodine species (Fig. 22k). XPS further confirms that Mo_4/3B₂T₂ achieves complete conversion from I₀ to I⁻ during discharge without detecting I₃⁻ formation, indicating its one-step I⁻/I₀ reversible conversion oxidation–reduction path, effectively avoiding polyiodide shuttle (Fig. 22l). Benefiting from the excellent electronic conductivity, effective iodine species confinement ability, and outstanding structural stability of the constructed battery system, the full battery exhibits nearly 100% CE and an exceptionally long cycle life, maintaining stable capacity output even under high current density. This provides a promising path for the development of high-performance aqueous Zn–I₂ batteries with ultra-long lifetimes and fast dynamic responses. These works highlight the unique advantages of MXenes and MBenes in regulating key interface behaviors in Zn–I₂ batteries. They not only achieve efficient confinement of iodine species but also synergistically enhance the cathode iodine loading capacity and anode reversibility, significantly suppressing polyiodide shuttle effects and the side reactions. This provides valuable guidance for designing new controllable host–guest structures based on 2D functional materials and developing high-stability zinc–halogen batteries.

Fig. 22 (a) SEM image and (b) AFM image of TMX. (c) UV-vis spectra of a mixture electrolyte of 2 M ZnSO₄ and 0.2 M ZnI₂ (ZSI) after charging to 1.8 V with TMX and CC as host materials, respectively. (d) Surface adsorption energy of TMX and CC toward I₃⁻ species. (e) Intermolecular distance variations of (i) CC and (ii) TMX upon I₃⁻ adsorption. (f) Corresponding contour plots of the charge density difference for I₃⁻ adsorption on (i) CC and (ii) TMX. (g) Visualization of the MD simulation model for ZSI-n and the corresponding enlarged view. (h) Radial distribution functions of the coordinated bond with Zn²⁺ in the ZSI-n system.¹³⁵ Reproduced with permission from ref. 135. Copyright 2024, Springer Nature. (i) Preparation process of Mo_4/3B₂T₂@I₂. (j) Electron density difference diagrams of I⁻, I₃⁻, and I₂ adsorbed on the Mo_4/3B₂T₂ MBene. (k) Comparative diagram of adsorption energy of I⁻, I₃⁻, and I₂ on Mo_4/3B₂T₂ MBene, Mo₂B₂T₂ MBene, and other reported host materials. (l) GCD curves of the Mo_4/3B₂T₂ electrode and the corresponding 3d XPS spectra at specific voltage states.¹³⁷ Reproduced with permission from ref. 137. Copyright 2023, Wiley-VCH.

3.1.5. Other materials. Typical host materials for halogen confinement primarily rely on van der Waals force-dominated physical adsorption, which exhibits weak interactions with various iodine species, making it difficult to achieve effective fixation during long-term cycling. As cycling progresses, polyiodide anions may gradually desorb from the host structure and re-dissolve into the electrolyte, causing rapid capacity decay of the Zn–I₂ battery. Therefore, it is necessary to develop advanced confinement hosts with strong host–guest interactions and chemical adsorption capabilities to enhance the performance of Zn–I₂ batteries. To address this issue, Zhang et al.¹³⁸ proposed the use of a natural, inexpensive biopolymer, starch, as a host material for efficient capture and stable confinement of polyiodide anions. Due to the unique double-helix structure of starch, it can form stable complexes with iodine, leading to strong light absorption. When starch chains come into contact with iodine, they exhibit a distinct blue–violet color change (Fig. 23a), indicating that starch can form strong bonds with polyiodide anions, effectively capturing them. UV-vis spectroscopy confirmed that starch has a good adsorption capacity for various iodine species, such as I⁻, I₃⁻, and I₂ (Fig. 23b). XPS depth profiling showed that in the starch/iodine composite, I₃⁻ primarily accumulates on the surface, and as the XPS etching process progresses, the dominant species transition to I₅⁻ (Fig. 23c), indicating that starch has some selectivity for confining higher-valent polyiodide anions. The comparison of surface and bulk species distribution (Fig. 23d) further reveals that the iodine species confined in starch are mainly I₅⁻. DFT results further show that as the structure evolves from monosaccharide units to a double helix, the binding energy between starch and iodine species decreases, indicating that the confinement ability gradually increases (Fig. 23e). Thanks to this strong confinement effect, starch-based Zn–I₂ batteries demonstrate excellent rate performance, significantly outperforming traditional Zn–I₂ batteries using KB as the cathode host (Fig. 23f). In situ Raman spectroscopy further revealed that I₃⁻ and I₅⁻ act as intermediates in the reversible conversion of I⁻/I₂ (Fig. 23g). During charging, the Raman signals of these two species first increase and then decrease, ultimately disappearing at high voltage intervals, indicating that they are fully oxidized to I₂. During discharge, the Raman signals also show an increase and then decrease, ultimately being fully reduced to I⁻, indicating that the battery system exhibits good reaction reversibility and enables efficient reversible conversion of I⁻/I₂. Therefore, the starch-based Zn–I₂ battery constructed using this structural confinement strategy demonstrates excellent electrochemical performance. Building on the significantly enhanced confinement ability of natural polymer systems, researchers have turned their attention to structurally adjustable organic functional polymers to achieve more precise regulation of iodine species and further optimization of electrode reaction processes. Recently, Zhang et al.¹³⁹ used in situ quaternization technology to prepare a unique N,N′-dimethy1-1,3-propylenediamine grafted and triethylenetetramine cross-linked polyacrylonitrile/iodine (GC-PAN/I) cathode (Fig. 23h). This material forms a uniform electrostatic confinement layer via a simple two-step in situ self-assembly process, stabilized by strong electrostatic interactions between the quaternary ammonium cation network and polyiodide anions, effectively enabling reversible conversion between I⁻ and I₃⁻ in solution. DFT calculations revealed that the GC-PAN/I material contains three types of functional sites: quaternary ammonium centers (site 1), amide centers (site 2), and secondary amine centers (site 3). Among these, the quaternary ammonium center is considered the key active site for adsorption and effective inhibition of I⁻ and I₃⁻ migration (Fig. 23i). COMSOL multiphysics simulations further suggested that the GC-PAN/I cathode does not generate solid I₂ during charging and discharging, and the reaction mainly proceeds via solution-phase conversion between I⁻/I₃⁻ (Fig. 23j and k), which was further verified by in situ UV-vis testing (Fig. 23l). Additionally, DFT calculations evaluated the energy barriers for the I⁻ → I₃⁻ → I₂ conversion at the three active sites, and the results showed that the conversion from I₃⁻ to I₂ has a high energy barrier at all three sites. This theoretical result confirms the experimental observation of the absence of solid I₂ and verifies that the GC-PAN/I cathode mainly undergoes the solution-phase I⁻/I₃⁻ reaction (Fig. 23m). Furthermore, the ordered 1D structure of the acrylonitrile fibers can significantly shorten the migration path of Zn²⁺ and improve iodine utilization. As a result, the GC-PAN/I-based Zn–I₂ battery demonstrates excellent cycling stability at 20 °C and high rates. In conclusion, non-framework organic polymers, with their highly designable structures, diverse polar functional groups, excellent iodine confinement capabilities, and good reversible redox activity, show great potential in Zn–I₂ batteries. However, these materials generally suffer from poor electrical conductivity and tend to swell or dissolve in aqueous environments, limiting their structural stability and cycling performance. Therefore, future research should focus on fine-tuning the material structure (e.g., introducing aromatic rings or cross-linked networks) and optimizing the electrical conductivity to further promote their practical applications in high-performance zinc–halogen batteries.

Fig. 23 (a) Schematic of the structure of starch and the starch/polyiodide complex. (b) UV-vis spectra of iodide/H₂O, triiodide/H₂O, and I₂/ethanol solutions before and after adsorption by starch. (c) I 3d XPS spectra of the starch/polyiodide complex. (d) Percentage of I₃⁻ and I₅⁻ species in the starch/polyiodide complex. (e) Comparative diagram of binding energies between I₂, I₃⁻, and I₅⁻ with starch monomers, six-membered rings, and double helices. (f) Rate performance of Zn–I₂ batteries with KB and starch cathodes at various current densities. (g) In situ Raman spectra of I⁻/I₂ conversion in starch-based Zn–I₂ batteries.¹³⁸ Reproduced with permission from ref. 138. Copyright 2022, Wiley-VCH. (h) Synthesis of GC-PAN/I and its advantages. (i) Binding energy of I⁻, I₃⁻, and I₂ at the quaternary ammonium center. Differences in I⁻ concentration distribution during discharge in a Zn–I₂ battery for two reaction pathways: (j) from I₃⁻ to I⁻ and (k) from I₂ to I⁻. (l) In situ UV-vis spectra of the GC-PAN/I cathode during the discharge process. (m) Oxidation energies of I⁻, I₃⁻, and I₂ at quaternary ammonium centers (site 1), amide centers (site 2), and secondary amine centers (site 3).¹³⁹ Reproduced with permission from ref. 139. Copyright 2022, Wiley-VCH.

As a nitrogen-rich aromatic heterocyclic molecule, 1,2,4,5-tetrazine (TTZ) has attracted significant attention due to its unique electronic structure and tunable chemical reactivity. TTZ molecules have low-energy π orbitals, which can stably form radical anions, exhibiting excellent reversible redox behavior.¹⁴⁰ Moreover, its highly functionalizable positions at the 3,6-positions allow it to build selective non-covalent interactions (such as hydrogen bonding, π–π stacking, electrostatic interactions, etc.) with specific guest molecules like polyhalide anions.¹⁴¹ Given these characteristics, TTZ is considered a strong candidate for constructing new materials for iodine species confinement and catalytic conversion, with potential for efficient capture and reversible conversion of polyhalide anions in zinc–halogen battery systems. Based on this, Qi et al.¹⁴⁰ proposed a molecularly engineered tetrazine derivative, 3,6-bis (2-morpholinoethyl)-1,2,4,5-tetrazine (BMT), as a multifunctional cathode host and applied it in Zn–I₂ batteries. BMT exhibits reversible two-electron redox characteristics. During charging, the reduced form of BMT, H₂BMT, can be oxidized back to BMT at a lower potential (−0.29 V) than the iodine redox reaction, preferentially participating in the energy storage process (Fig. 24a). DFT calculations show that the LUMO of BMT is located in the tetrazine unit, while the HOMO of H₂BMT is concentrated in the reduced dihydrotetrazine portion, clarifying the intrinsic mechanism of its two-electron transfer (Fig. 24b). Further ESP mapping analysis reveals that the reduced dihydrotetrazine fragment has enhanced nucleophilicity (Fig. 24c). This evidence suggests that BMT can effectively suppress the shuttle behavior of polyiodide anions through a covalent-electrostatic synergistic confinement mechanism, thereby stabilizing the iodine redox conversion process. In a 2 M ZnSO₄ + 0.2 M KI electrolyte system, BMT-based batteries retain a good capacity retention rate after 3500 cycles, maintaining a stable I⁻/I₃⁻ potential plateau, fully verifying BMT's regulatory and stabilizing effect on iodine redox behavior (Fig. 24d). In situ Raman tests further reveal BMT's confinement effect on iodine species. In the ZnSO₄/KI electrolyte, compared to the AC electrode, which shows distinct I₃⁻ and I₅⁻ vibration peaks at 1.6 V (Fig. 24e), the BMT electrode initially exhibits fluorescence signals from the TTZ unit's redox at 1.1 V, with weak iodine-related peaks, indicating that it prioritizes energy storage during the initial charging phase and can effectively bind with nascent I₃⁻, limiting its migration in the electrolyte (Fig. 24f). Geometric structure analysis shows that linear I₃⁻ binds to the BMT's TTZ unit at a 90° angle, with charge distribution forming a negative electrostatic center in the TTZ region (Fig. 24g). ¹³C NMR analysis shows significant high-field shifts in all signal peaks upon adding 2 molar equivalents of TBAI₃ to the BMT solution, indicating an increase in electron cloud density and exhibiting distinct anion–π interaction characteristics (Fig. 24h). This demonstrates that BMT can also form stable precipitates with I₃⁻ ions at a 1 : 2 stoichiometric ratio, effectively suppressing polyiodide shuttle through covalent–electrostatic synergistic confinement. FTIR tests further support the formation of these precipitates (Fig. 24i). Thanks to this two-electron redox reaction and covalent-electrostatic adsorption mechanism, BMT-based cathode materials achieve extremely high iodine anchoring efficiency and CE, with a super-long cycle life of over 33 000 cycles in coin cells. Furthermore, to address the slow redox activity and effective iodine confinement issues in both carbon-based and organic materials, Zhang et al.¹⁴² combined carbon-based materials with organic materials to construct a nitrogen-modified alternating copolymer nanoflower composite material (NH-NFs-PDA, where PDA stands for polydopamine) as a high-performance iodine host material in Zn–I₂ batteries (Fig. 24j). This host material enhances the adsorption of polyiodide anions and the ion transport rate through the design of high-density hydroxyl groups in the alternating copolymer backbone, the introduction of nitrogen elements, and the porous nanostructure. UV absorption results show that, in the battery system with NH-NFs-PDA/I₂ as the cathode, the absorption intensity of I₃⁻ in the electrolyte remains low and nearly unchanged during charging, indicating that polyiodide shuttle is effectively suppressed (Fig. 24k). DFT calculations reveal that the higher electronegativity of nitrogen atoms in the (N-modified polyhydroxy alternating copolymer) NH–P backbone of NH-NFs-PDA/I₂ enhances the inductive effect on the electron cloud of hydroxyl oxygen atoms, increasing the positive polarity of the hydroxyl hydrogen atoms, thereby enhancing the binding ability with I₃⁻ (Fig. 24l). Benefiting from this structural design, the Zn–I₂ battery constructed with NH-NFs-PDA/I₂ demonstrates excellent rate performance and an ultra-long cycle life, showcasing the immense potential of alternating copolymers in building multifunctional polymer iodine hosts.

Fig. 24 (a) Reaction mechanism of Zn–I₂ batteries with bifunctional BMT as a cathode. (b) Redox reaction mechanism of BMT in a Zn–I₂ battery. (c) ESP mapping of H₂BMT and BMT. (d) GCD curves of the BMT cathode at a current density of 2 A g⁻¹. CV curves of the (e) AC cathode and (f) BMT cathode, along with their corresponding in situ Raman spectra. (g) Job's plot profiles of the BMT–I₃⁻ complex and ESP mapping of the 2I₃⁻@BMT complex. (h) ¹³C NMR spectra of 2I₃⁻@BMT and BMT. (i) FTIR spectra of TBAI₃, 2I₃⁻@BMT and BMT.¹⁴⁰ Reproduced with permission from ref. 140. Copyright 2025, Wiley-VCH. (j) Schematic diagram of the synthesis process of NH-NFs-PDA/I₂ and the assembly of the Zn–I₂ battery, and the interaction between NH-NFs-PDA and iodine species. (k) In situ UV-vis spectra of the electrolyte during charging with NH-NFs-PDA/I₂ as the cathode. (l) Binding energies of I₃⁻ at the hydroxyl sites of PDA, the alternating copolymer used in SH-NFs-PDA (SH-P), and NH-P.¹⁴² Reproduced with permission from ref. 142. Copyright 2025, Wiley-VCH.

Aqueous-based photo-assisted batteries have garnered widespread attention due to their unique advantages in integrated solar energy conversion and storage.¹⁴³ However, finding an ideal photoelectric cathode that can simultaneously capture iodine and exhibit photo-responsive behavior remains a major challenge for the development of photo-assisted Zn–I₂ batteries. Therefore, Xu et al.¹⁴⁴ designed a BiOI material with abundant iodine vacancies to achieve enhanced energy storage behavior through synergistic iodine confinement and photoelectric response (Fig. 25a). Electrochemical tests combined with in situ Raman results show that this material can achieve highly reversible iodine redox processes, primarily relying on two energy storage pathways: first, iodine vacancies in the BiOI crystal confine the adsorption and reversible conversion of I⁻, corresponding to reversible switching between the Bi–O and I–Bi–O configurations; second, the two-step redox process of iodine, in which solid I₂ is first generated, followed by further oxidation to form polyiodide anions dominated by I₅⁻ (Fig. 25b–d). In situ Raman spectroscopy of BiOI reveals that the I₅⁻ signal exhibits reversible generation and disappearance during charging and discharging, indicating effective anchoring by BiOI. Representative in situ Raman spectra at different charge and discharge states (Fig. 25e) further confirm that I⁻ in BiOI first oxidizes to form solid I₂, which then further transforms into I₅⁻, corresponding to the two-step reversible redox process. The O₂/R₁ peak is attributed to the generation and reduction of I₂, while the O₃/R₂ peak corresponds to the reversible conversion of I₅⁻. Additionally, constant potential tests clearly show that the BiOI material can rapidly generate photogenerated electron–hole pairs under sunlight under illumination. The electrons migrate through the external circuit to the anode, promoting the reduction of Zn²⁺, while the holes at the electrode surface facilitate the migration and oxidation of I⁻ in the electrolyte, creating an additional energy storage channel involving photogenerated carriers, as shown in the mechanism diagram (Fig. 25f). Therefore, the BiOI cathode maintains a relatively high mid-discharge voltage at various rates, significantly outperforming traditional carbon materials (Fig. 25g). The photo-assisted Zn–I₂ battery assembled with BiOI material demonstrates excellent energy storage performance at a current density of 1 mA cm⁻² and shows a significant capacity improvement under illumination, exhibiting a clear photo-response enhancement effect (Fig. 25h). Flexible device tests further prove that it maintains stable performance under various bending angles, enabling integration applications such as powering smartphones and wearable health monitoring devices (Fig. 25i and j). This study demonstrates the unique advantages of BiOI-based photoelectric cathodes in achieving efficient iodine species regulation and synergistic photo-responsive energy storage, providing strong support for the practical exploration of photo-assisted water-based Zn–I₂ batteries. These studies confirm that organic polymers and novel topological insulators can significantly enhance iodine species (such as I⁻, I₃⁻, and I₅⁻) confinement and anchoring effects by constructing specific chemical bonding sites, introducing covalent-electrostatic synergistic interactions, or achieving energy level regulation. This effectively suppresses shuttle behavior, improves CE, and enhances cycling stability. This drives the development of the design concept of the confinement–catalysis–conduction for efficient host materials.

Fig. 25 (a) Diagram of a flexible Zn–I₂ battery with a BiOI cathode. (b) CV curves at 0.2 mV s⁻¹, (c) GCD curves at 1 mA cm⁻², (d) in situ Raman spectra, and (e) representative Raman spectra of BiOI during the charge–discharge process. (f) Energy storage characteristics of BiOI in photo-assisted Zn–I₂ batteries. (g) Ratios of Bi–O and I–Bi–O of BiOI at different charge–discharge states. (h) GCD curves of the wearable device composed of the BiOI cathode, PAM hydrogel electrolyte, and zinc anode under light and dark conditions at a current density of 1 mA cm⁻². (i) Electrochemical stability of the flexible photo-assisted Zn–I₂ battery under different bending angles. (j) Application of the flexible device in wearable health monitoring.¹⁴⁴ Reproduced with permission from ref. 144. Copyright 2024, Wiley-VCH.

3.2. Halogen-confining host materials for Zn–Br₂ batteries

Zn–Br₂ batteries have emerged as promising candidates for next-generation energy storage systems owing to their low cost, non-flammable aqueous electrolytes, deep discharge capability, long cycle life, and excellent redox reversibility.¹⁴⁵ However, in the absence of effective Br host materials, the Br₂ generated during charging readily reacts with Br⁻ to form soluble polybromide ions (Br_n⁻, n = 3, 5, 7), which can shuttle between electrodes, leading to severe self-discharge, Zn anode corrosion, and reduced CE.¹⁴⁶ To address these challenges, the development of advanced Br hosts capable of effectively confining Br species and enhancing redox reversibility is essential. In recent years, a variety of host materials including carbonaceous materials, functional frameworks and their derivatives, as well as other cathode host systems have been extensively investigated due to their tunable porosity, abundant anchoring sites, and modifiable surface chemistry, enabling efficient Br immobilization, suppression of shuttle effects, and improved cycling performance in Zn–Br₂ systems. A summary of representative studies in this section is provided below.

3.2.1. Carbonaceous materials. Carbonaceous materials offer several advantages, including high chemical stability, excellent electrical conductivity, tunable surface properties, and strong resistance to corrosive media, making them ideal host materials for bromine cathodes in Zn–Br₂ batteries.¹⁴⁷ With the industrial applications of Zn–Br₂ batteries, carbon-based electrodes are gaining considerable attention. For example, Wang et al.¹⁴⁸ investigated four representative commercial carbon materials – acetylene black (AB), expanded graphite (EG), CNT, and BP2000 (BP) – to systematically explore the intrinsic relationship between the structural characteristics of carbon materials and the Br₂/Br⁻ electrochemical activity in Zn–Br₂ batteries. N₂ adsorption–desorption experiments were conducted to compare the pore structures of different materials. The results show that the electrochemical activity of carbon materials generally correlates positively with their specific surface area. Notably, the CNT exhibited a significantly better performance than expected based on its specific surface area, which was attributed to its excellent conductivity. Furthermore, an appropriate pore size distribution, abundant macropores, and numerous micropores effectively promoted mass transfer, while the high degree of graphitization led to excellent conductivity that greatly reduced ohmic resistance, enhancing electrochemical performance. However, excessive graphitization reduced specific active sites and slowed down reaction kinetics. Therefore, further systematic studies on the effects of pore size distribution and internal structure of carbon materials on bromine species confinement will help guide the rational design of high-performance Zn–Br₂ battery electrodes. Recently, Wang et al.¹⁴⁹ used the SiO₂ template method combined with CO₂ activation to successfully construct a cage-like porous carbon (CPC), achieving efficient confinement of bromine species by controlling its shell pore size and internal structure (Fig. 26a). DFT calculations show that the pore size of CPC is between that of Br⁻ and the Br₂ complex (MEPBr₃), allowing Br⁻ to freely enter the cage, be oxidized to Br₂ inside the cavity, and form a complex with MEP⁺ to generate MEPBr₃, which is ultimately stabilized in the cage due to the pore size sieving effect (Fig. 26b). In contrast, solid carbon spheres (CS) and hollow carbon spheres (HCS) have some structural characteristics but fail to effectively restrict Br₂ diffusion due to the lack of confinement channels in the shell (Fig. 26c and d), demonstrating the unique structural advantages of CPC. Additionally, the large number of active sites formed during CO₂ activation further accelerated the Br₂/Br⁻ redox reaction rate, enhancing the electrode activity while effectively anchoring the active species. As a result, Zn–Br₂ flow batteries based on CPC achieved a CE of up to 98% and an energy efficiency of 81% at a current density of 80 mA cm⁻². This work, through the construction of reasonable confinement spaces, realized precise restriction and efficient fixation of Br₂ complexes, advancing the development of long-life, high-efficiency, and high-energy-density Zn–Br₂ batteries. However, due to the relatively poor hydrophilicity of some carbonaceous materials, such as carbon felts (CFs) and graphite felts (GFs), the kinetics of the Br₂/Br⁻ redox reaction are slow, and further research on CFs and GFs is required. Recent studies suggest that combining carbon materials with metal halides is an effective strategy to improve Br₂/Br⁻ redox reaction kinetics. For example, composites of zinc halides (such as ZnCl₂, ZnBr₂, and ZnI₂) with carbon materials (such as graphite or activated carbon), combined with Zn²⁺-conducting water-in-salt gel electrolytes, can realize an energy storage mechanism that does not rely on Zn²⁺ intercalation reactions.¹⁵⁰ In this system, halide ions not only serve as charge carriers but also directly participate in the redox reactions, effectively avoiding the slow kinetics associated with the strong binding of Zn²⁺ with the host lattice in traditional zinc-ion batteries. Unlike conventional Zn–Br₂ flow batteries, which store Br₂ in external tanks, this strategy fixes the zinc halides within the carbon host, allowing Br₂ to undergo a “conversion-intercalation/adsorption-conductivity” process within the carbon structure, significantly suppressing the halogen shuttle effect and enhancing electrode activity and energy density.

Fig. 26 (a) Diagram of the synthesis of cage-like porous carbon. (b) Br⁻ confinement mechanism in cage-like porous carbon. SEM images of CS (c) and HCS (d).¹⁴⁹ Reproduced with permission from ref. 149. Copyright 2017, Wiley-VCH. (e) Schematic of the CARCO platform composed of transformer-based language models, robotic CVD, and data-driven ML.¹⁵¹ Reproduced with permission from ref. 151. Copyright 2025, Elsevier.

However, traditional material design methods based on empirical experience have gradually shown limitations in terms of efficiency and accuracy. Recently, AI technology has demonstrated strong auxiliary potential in material synthesis and structural optimization. For example, Li et al.¹⁵¹ developed an intelligent platform, CARCO, which integrates three core modules: customized language models (such as Carbon_GPT and Carbon_BERT), an automated chemical vapor deposition (CVD) system, and ML algorithms trained on real experimental data, to explore the synthesis strategies for high-density, horizontally aligned carbon nanotube (HACNT) arrays (Fig. 26e). Using this platform, they successfully selected a novel titanium–platinum bimetallic catalyst, achieving efficient and controlled growth of the HACNT arrays. This intelligent synthesis paradigm not only significantly enhances the structural design efficiency and performance regulation precision of carbon-based materials but also offers new solutions for material synthesis under multivariable conditions, demonstrating broad potential for expansion in Zn–Br₂ batteries.

3.2.2. Functional frameworks. Functional frameworks, such as MOFs, COFs, and PBAs, possess highly ordered pore structures and large specific surface areas. These materials, with their high specific surface area, rich pore structures, and tunable chemical properties,¹⁵² can effectively suppress the loss and volatilization of bromine species through a synergistic effect of physical adsorption and chemical anchoring. This provides an ideal platform for the efficient storage and stable operation of bromine species. The mechanism of anchoring bromine species depends on the design characteristics of the framework material. On one hand, the pore structure provides spatial constraints for the physical adsorption of bromine species,¹⁴⁹ while on the other hand, functional groups introduced into the framework, including nitrogen groups, oxygen groups, or metal coordination sites, can interact chemically with bromine species, anchoring them firmly through the formation of hydrogen bonds, coordination bonds, or charge transfer interactions.¹⁵³ This dual action mechanism not only increases the storage capacity of bromine species but also significantly inhibits the occurrence of zinc anode side reactions, ultimately achieving better electrochemical performance in Zn–Br₂ batteries. Recently, researchers have begun exploring material systems that combine both adsorption and catalytic functions. For example, Wei et al.²⁸ reported a novel 2D conjugated nickel phthalocyanine (NiPPc) material, which, as an adsorption–catalysis dual-functional cathode host, demonstrates excellent performance (Fig. 27a). NiPPc has a highly conjugated 2D layered structure and rich porosity, providing good polarity and electron transport capabilities. Its atomically dispersed Ni–N₄ coordination structure not only provides stable anchoring sites but also serves as an efficient catalytic center for bromine oxidation–reduction reactions (Fig. 27b and c). To further investigate the Br⁰/Br⁻ reversible redox mechanism and the interaction between NiPPc and bromine species, the researchers conducted in situ UV-vis spectroscopy during the battery's charge and discharge cycles. The results showed that as the voltage gradually decreased, the absorption peak intensity of polybromine compounds in the solution continued to weaken and nearly disappeared at full discharge, indicating that polybromine species underwent reversible generation and consumption during the charge–discharge process (Fig. 27d), confirming their participation in electrochemical reactions and being controlled by the anchoring and regulation of the NiPPc structure. Furthermore, the XPS fitting peaks of Br⁻, Br–Br, and Br–O/Br–C species observed in the tests further confirmed that bromine underwent a stepwise conversion from elemental bromine to polybromides at the electrode surface (Fig. 27e), and in situ Raman tests indicated that the Ni–N₄ centers in NiPPc provided effective catalytic sites for the Br₂/Br⁻ redox reaction (Fig. 27f). Theoretical calculations further revealed that this material exhibited strong adsorption energy during the adsorption of Br⁻, Br₂, and Br₃⁻, and the differential charge density and Bader charge analysis showed significant interfacial charge redistribution during the adsorption of bromine species, indicating that NiPPc could effectively capture and confine active intermediates, enabling molecular-level regulation of bromine species (Fig. 27g). Notably, in the free energy calculations for halogen evolution reactions, NiPPc showed lower reaction free energies at each intermediate state compared to control materials without metal coordination structures, indicating that it could effectively reduce reaction barriers and enhance the kinetics of the bromine oxidation–reduction reactions (Fig. 27h). Thanks to the synergistic advantages of physical confinement, chemical anchoring, and electrocatalysis, NiPPc significantly enhances the discharge voltage platform, capacity retention, and cycling stability of the battery, providing a novel material design strategy for the controllable conversion of bromine species and the development of high-performance water-based Zn–Br₂ batteries. In addition to the excellent confinement and catalytic performance exhibited by transition metal coordination structures, scientists have recently emphasized the potential of exfoliated 2D COFs with single or few-layer structures, which effectively shorten the charge migration distance. For example, Zhang et al.³⁴ successfully prepared a 2D ketone-amine bonded covalent organic framework (exCOF) as a cathode host material for Zn–Br₂ batteries. The exfoliated porous COF structure exposes abundant active sites, effectively stabilizing polybromides and facilitating their conversion, thereby significantly suppressing the shuttle effect of polybromides. Research shows that the large number of oxygen-containing functional groups introduced into the exCOF can interact strongly with bromine and polybromide anions, achieving efficient confinement and adsorption of active species, thereby stabilizing the presence of polybromine species and promoting their reversible conversion during charging and discharging (Fig. 27i). DFT calculations also show that the C=O groups in the exCOF exhibit optimal adsorption energy for bromine species, and there is significant charge exchange between the functional groups and bromine species, further demonstrating the chemical anchoring effect at the interface (Fig. 27j). Leveraging the confinement and electrocatalytic conversion abilities of exCOF towards bromine species, the Zn–Br₂ battery constructed with this material as the cathode exhibits excellent electrochemical performance. To further expand the application potential of COFs in confinement regulation, many studies have started focusing on the effects of their structural evolution and microstructural features during the construction process on performance. Zheng et al.¹⁵⁴ polymerized COF precursors into nanoporous sheet units and further assembled them into COF nanostructures with “onion-like” concentric layered arrangements. Using advanced liquid-phase TEM (liquid-phase TEM) and cryo-TEM, they achieved real-time visual observation of the structural construction process of these COF nanostructures (Fig. 27k). This experimental system, when cooled to −196 °C with liquid nitrogen, effectively avoided radiation damage to the beam-sensitive COF material by high-energy electron beams, enabling high-resolution imaging of its local fine structure. During in situ imaging, the researchers successfully captured multiple key intermediate steps, including graphite sheet formation, interlayer adhesion, interlayer spacing relaxation, and structural homogenization (Fig. 27l), providing a comprehensive revelation of the dynamic evolutionary path of these multilayer COF nanostructures growing stepwise in solution. This work not only deepens the understanding of the structural construction mechanism of the COF nanostructure but also lays the foundation for exploring the confinement–catalysis–conduction design of host materials.

Fig. 27 (a) Working principle of the NiPPc host applied in aqueous non-flow Zn–Br₂ batteries. (b) FT-EXAFS spectra of NiPPc and Ni-foil*0.3. (c) Fitting curves of NiPPc EXAFS spectra. (d) UV-vis spectra of the electrolyte soaked with the KBr-NiPPc electrode at different charge–discharge voltages. (e) Br 3d XPS spectra of KBr-NiPPc in the pristine state and at specific charge–discharge voltages. (f) In situ Raman spectra of bromine species at the electrode–electrolyte interface. (g) Adsorption energy of Br⁻, Br₂, and Br₃⁻ on Ni–N₄ and NiPPc rings, along with the corresponding differential charge density maps after adsorption. (h) Gibbs free energy profiles of the bromine reduction reaction (BrRR) process on polyphthalocyanine (PPc) and NiPPc.²⁸ Reproduced with permission from ref. 28. Copyright 2023, Royal Society of Chemistry. (i) GCD curves of the Br₂-exCOF cathode and the corresponding in situ Raman spectra. (j) Adsorption energy of Br⁻, Br₂, and Br₃⁻ on N–H, N=N, and C=O sites, along with the charge density patterns at the C=O adsorption site.³⁴ Reproduced with permission from ref. 34. Copyright 2023, Elsevier. (k) Schematic diagram of the combined method integrating liquid cell technology with cryo-electron microscopy. (l) HRTEM images of the COF onion nanostructure.¹⁵⁴ Reproduced with permission from ref. 154. Copyright 2024, American Chemical Society.

3.2.3. Derivatives of functional frameworks. In recent years, MOFs have been widely used as important precursors for constructing ordered carbon materials due to their high specific surface area, rich metal and organic components, large pore volumes, and high tunability in structure and composition.¹⁵⁵ However, the preparation of porous carbon based on MOFs typically requires the introduction of small molecular carbon sources with high vapor pressure into their pores, which complicates the process and results in limited filling efficiency.¹²³ Additionally, most MOF materials are environmentally sensitive, have poor structural stability, and exhibit low yields, which severely restricts their scalability for applications.¹⁵² Therefore, there is an urgent need to develop an alternative strategy that is simple, cost-effective, and scalable. For example, Wang et al.¹⁵⁶ synthesized a nanosheet-type zeolite imidazolate framework (NSZIF) by stirring zinc acetate and 2-methylimidazole in an aqueous solution for 24 hours. They then pyrolyzed it under a CO₂ atmosphere to prepare a porous nanosheet carbon material (PNSC) with a planar pore structure, which was used as the cathode material for Zn–Br₂ batteries. Such a unique nanosheet morphology helps shorten the electron transport path and exhibits excellent conductivity. Its high porosity, planar nanopores, and loose structure provide a larger interface for electrolyte contact, while offering more active sites for Br₂/Br⁻ electrochemical reactions. Moreover, PNSC is rich in electronegative nitrogen–oxygen functional groups, which effectively adsorb Br⁻ and Br₃⁻ anions and form adsorbed bromine atoms or Br₂ molecules through ion exchange. This mechanism significantly improves the host material ability to capture bromine species and enhance its electrochemical activity, allowing PNSC to demonstrate excellent performance in Zn–Br₂ flow batteries. As research into derivatives of functional frameworks progresses, the diversity and tunability of these materials will offer new opportunities to further enhance the performance of Zn–Br₂ batteries.

3.2.4. 2D transition metal carbides/nitrides. 2D transition metal carbides/nitrides, such as MXenes, have unique advantages in electrode design due to their excellent conductivity, periodic layered microstructure, array-like interlayer gaps, rich active metal atoms, and surface functional groups (=O, –F, –OH, etc.).³² The MXene skeleton efficiently conducts electrons and provides support to improve the performance of insulating active materials, enabling ultrafast reaction kinetics.¹⁵⁷ Meanwhile, its nanoscale interlayer channels limit the diffusion of active substances and inhibit common shuttling behaviors. Additionally, the ceramic properties of MXenes effectively buffer volume changes, providing good structural stability.¹⁵⁸ Researchers have thus explored using MXenes as halogen-confining host materials for zinc–halogen batteries to limit the dissolution and shuttling effects of halogen species and promote efficient redox reactions. Studies have confirmed that the polar transition metal layers exposed with O/F functional groups have a good affinity for polyhalides, which helps suppress the loss and shuttle behavior of active reaction products. However, considering the large radius of the Br₂ molecule and the nanoscale interlayer spacing of MXenes, traditional methods such as physical adsorption and liquid-phase infiltration are inefficient for achieving a high degree of the interphase composite. To address this, Li et al.¹⁵⁹ proposed an effective “MXene-enhancement” strategy, where Br₂ is uniformly embedded into the interlayer space of Ti₃C₂T_x nanosheets through an electrochemical deposition method (Fig. 28a). Under a constant electric field, Br⁻ ions first insert into the MXene interlayer and are in situ oxidized to elemental Br₂, achieving stable confinement of bromine species. The structural evolution and interfacial behavior of bromine in the electrode were verified through systematic tests: XRD results showed that the Ti₃C₂T_x crystal structure remained stable at different voltage states, with the (002) crystal plane experiencing only slight low-angle shifts during charging and returning to the original position during discharge (Fig. 28b). The Br 3d XPS fitting results further confirmed that bromine species underwent reversible oxidation–reduction reactions during the charge–discharge process (Fig. 28c). Meanwhile, DFT calculations showed that Ti₃C₂T_x exhibits thermodynamic spontaneity for the adsorption of Br⁻, Br₂, and Br₃⁻, especially on Ti₃C₂T_x surfaces rich in O-functional groups, where bromine species adsorb most stably at Ti sites (Fig. 28d and e). The charge differential density map shows significant charge redistribution in the interface region during adsorption at Ti sites (Fig. 28f), indicating significant electron exchange between the active bromine species and Ti₃C₂T_x, facilitating the redox reaction kinetics. Furthermore, its interlayer nanospace provides effective physical confinement for bromine species such as Br⁻, Br₂, and Br₃⁻, inhibiting their migration and leakage, thus alleviating the shuttle effect. Thanks to this enhanced confinement mechanism, Br–Ti₃C₂T_x electrodes demonstrated a stable discharge platform at 1.75 V, excellent rate performance, and cycle life in aqueous Zn electrolytes. However, traditional etching methods for preparing MXenes typically require hydrofluoric acid, posing significant environmental and safety risks. To address this, Guo et al.¹⁶⁰ employed a green molten salt method to successfully synthesize highly conductive Ti₂CT_x materials (Fig. 28g). The abundant Ti active sites and O-terminal functional groups on the Ti₂CT_x surface effectively adsorb dissolved polybromide anions, significantly suppressing the shuttle effect, reducing active material loss, and enhancing CE and cycling stability. Compared to traditional carbon-based materials, which mainly rely on physical confinement and exhibit weak adsorption of polar bromine anions, Ti₂CT_x not only has a stronger chemical adsorption ability but also possesses excellent catalytic activity, accelerating the reversible conversion of polybromide anions and significantly improving reaction kinetics (Fig. 28h). Therefore, the Zn–Br₂ battery assembled with Ti₂CT_x-based cathodes achieved over 3000 stable cycles and excellent capacity retention under non-flow, membrane-free conditions, significantly outperforming traditional carbon host systems. Moreover, to further alleviate the diffusion and shuttle problems of bromine species, bromine complexing agents (BCAs) were introduced into the electrolyte to optimize the electrode–electrolyte interface and enhance the regulation of active species. However, BCAs form oily polybromide complexes, reducing the bromine concentration in the aqueous phase and degrading the conductivity and homogeneity of the electrolyte, thereby deteriorating the redox kinetics of the Br₂/Br⁻ redox couple. To balance the pros and cons of BCA introduction and effectively mitigate the diffusion and shuttle effect of bromine, Tang et al.⁵¹ introduced cetyltrimethylammonium bromide (CTAB) as a BCA onto a Ti₃C₂T_x MXene substrate to combine the advantages of both BCAs and the MXene. Based on the strong and reversible solid complexation effect, bromine species are trapped/retained within the Ti₃C₂T_x MXene (Fig. 28i), effectively suppressing bromine diffusion and shuttling. In addition, the bromine adsorption ability of the Ti₃C₂T_x MXene accelerates the redox kinetics of the Br₂/Br⁻ redox couple. Therefore, the Zn–Br₂ battery assembled with a Ti₃C₂T_x-CTAB-modified cathode achieved ultrahigh capacity retention, high voltage efficiency, and high energy efficiency. To further promote the structural optimization and performance enhancement of MXene-based electrode materials, the emerging AI-assisted material design methods in recent years have provided new technological pathways for their applications. For instance, Li et al.¹⁶¹ achieved reverse design of MXene materials via multi-objective ML methods, predicting and optimizing key electrochemical properties such as specific capacity, voltage, and charge induction. By combining regression and classification models, they established the structure–performance correlation of MXenes and found that intercalation ions have the most significant impact on performance, followed by transition metals, surface functional groups, and layered substrates. The model further predicted that Sc-, Ti-, and Cr-based MXenes with Li⁺ and Mg²⁺ intercalation exhibit excellent performance and are worth considering as candidate materials. Based on these ML strategies, future research is expected to efficiently screen MXene materials with high-polarity surfaces, further enhancing their confinement and catalytic effects on polybromide anions, thereby significantly advancing the development of cathode materials for high-performance Zn–Br₂ batteries.

Fig. 28 (a) Diagram of the synthesis of Br–Ti₃C₂T_x. (b) XRD patterns and (c) Br 3d XPS spectra of the Br–Ti₃C₂T_x cathode at specific charge–discharge voltages. (d) Crystal structure of the Ti₃C₂T_x MXene in the top view. (e) Adsorption energy of Br⁻, Br₂, and Br₃⁻ on the Ti₃C₂T_x MXene and proposed possible active sites. (f) Charge density patterns of Br⁻, Br₂, and Br₃⁻ around the Ti adsorption sites.¹⁵⁹ Reproduced with permission from ref. 159. Copyright 2021, American Chemical Society. (g) Fabrication process of Ti₂CT_x. (h) Comparison of the defects and advantages of traditional carbon materials and the Ti₂CT_x MXene as bromine hosts.¹⁶⁰ Reproduced with permission from ref. 160. Copyright 2024, Wiley-VCH. (i) Working mechanism of a Ti₃C₂T_x-CTAB modified electrode applied in bromine-based flow batteries.⁵¹ Reproduced with permission from ref. 51. Copyright 2024, Royal Society of Chemistry.

3.2.5. Other materials. In addition to carbonaceous materials, functional frameworks and their derivatives, and MXene materials, non-framework organic polymers have also begun to be explored and applied in Zn–Br₂ batteries. Recently, Wu et al.¹⁶² selected an anion-exchange polymer with polybipyridine (PBP) as the main chain, and introduced hydrophilic (2-bromoethyl) trimethylammonium bromide (BTAB) and hydrophobic hexyl bromide (HB) onto the side chains to construct a new cathode for non-flow aqueous Zn–Br₂ batteries. This polymer possesses selective transport capability for bromine ions and limits the migration of polybromide anions through steric hindrance effects, reducing their dissolution and diffusion in the electrolyte. Additionally, based on the polymer's solubility in specific solvents, the researchers combined it with conductive carbon materials (SP carbon) to create a uniformly coated composite electrode (SP-PBH), achieving cooperative conduction of electrons and bromine ions. Testing showed that the polymer uniformly coated the carbon particle surface (Fig. 29a and b). The Br 3d XPS spectra indicated that the bromine species on the electrode surface in the charged state were present in the form of Br₅⁻ and Br₃⁻ polybromides, and after discharge, predominantly Br⁻ ions, confirming the reversible oxidation–reduction process between bromine ions and polybromides in the electrode (Fig. 29c and d). In situ UV-vis spectroscopy did not show any additional absorption peaks in the solution except for ZnBr₂, confirming that the polybromide ions were successfully confined within the SP-PBH composite (Fig. 29e). Thanks to the efficient bromine ion transport and effective confinement of polybromide anions provided by this polymer–carbon composite electrode design, the Zn–Br₂ battery based on the SP-PBH composite cathode overcame the performance degradation caused by the dissolution and diffusion of polybromide anions in traditional Zn–Br₂ batteries. It exhibited excellent low-temperature performance and stable cycling ability at 0 °C (Fig. 29f–h). Further assembled pouch cells also showed high CE, high energy efficiency, and good cycling stability at room temperature (Fig. 29i–k). Moreover, topological insulator materials (such as Bi₂Se₃, Bi₂Te₃, and their surface-coated Bi thin-film composites) also exhibit excellent adsorption performance for Br₂ and Cl₂.¹⁶³ Zhu et al.¹⁶⁴ systematically studied the interaction of these materials with halogen gases under ultra-high vacuum conditions and analyzed the surface composition and structural evolution before and after halogen adsorption. The results showed that Br₂ could undergo weak chemical adsorption on the Se or Te ends of Bi₂Se₃ and Bi₂Te₃ surfaces, with the adsorbed state being relatively unstable; however, after surface coating with Bi thin films, stronger adsorption and reactivity were observed, enabling reversible modulation of the surface structure. This controllable response to halogen species holds promise for effectively confining polybromide anions, inhibiting shuttle effects and improving the reversibility and cycling stability of redox reactions. Meanwhile, their layered structure and excellent electronic conductivity facilitate the construction of efficient electrode interfaces, showing broad application prospects in Zn–Br₂ battery systems. However, these materials still face several challenges in practical applications, including the difficulty of scalable synthesis due to complex structural design, inadequate electrode preparation process adaptability, and structural stability issues under long-term operation,¹⁶⁵ which require further research to promote their commercialization in high-performance Zn–Br₂ batteries.

Fig. 29 (a) SEM images of the SP-PBH composite and the corresponding energy dispersive X-ray spectroscopy (EDS) spectra. (b) HRTEM image of SP-PBH. Br 3d XPS patterns of the SP-PBH cathode under charge (c) and discharge (d) states. (e) In situ UV-vis spectra of the electrolyte with SP-PBH as the electrode at different charge and discharge voltages. (f) GCD curves and (g) energy efficiency, voltage efficiency and CE of the Zn–Br₂ battery with the SP-PBH composite electrode at different current densities at 0 °C. (h) Cycling performance at 8 mA cm⁻² of the Zn–Br₂ battery with the SP-PBH cathode at 0 °C. (i) Diagram of the pouch cell assembled with the SP-PBH cathode, along with its corresponding (j) GCD curves and (k) cycling performance.¹⁶² Reproduced with permission from ref. 162. Copyright 2022, Elsevier.

3.3. Halogen-confining host materials for Zn–Cl₂ batteries

Aqueous Zn–Cl₂ batteries, as a crucial subclass of zinc–halogen systems, have attracted growing attention owing to their high theoretical capacity, elevated operating voltage, low cost, and the natural abundance of chlorine resources, making them promising candidates for large-scale energy storage applications.⁵² However, their practical deployment is hindered by several key challenges. The oxidized species Cl₂ is highly unstable in aqueous environments, readily undergoing disproportionation or volatilization, and the Cl₂/Cl⁻ redox reaction suffers from sluggish kinetics, which severely compromises the cycling stability and CE.¹⁶⁶ In recent years, considerable efforts have been devoted to the development of halogen-confining host materials to mitigate these issues. In particular, carbonaceous materials such as AC, graphite, and CNTs have demonstrated the ability to physically confine oxidized chlorine species via pore confinement and adsorption interactions.¹⁶⁷ These host materials effectively suppress the shuttle effect and volatilization loss of Cl₂ while enhancing the reversibility of redox reactions, thereby significantly improving the CE and cycling life of Zn–Cl₂ batteries. This section focuses on representative confinement strategies and their underlying mechanisms in Zn–Cl₂ batteries.

3.3.1. Carbonaceous materials. The application history of the Cl₂/Cl⁻ redox couple in energy storage can be traced back to the 19th century. However, the oxidation product Cl₂ is prone to undesirable disproportionation reactions in the electrolyte, which reduces the utilization efficiency of chlorine species.¹⁶⁸ Currently, there are few research reports on Zn–Cl₂ batteries, and they are limited by a series of technical challenges in practical applications. Aqueous Zn–Cl₂ batteries typically use carbonaceous materials as cathode host materials. For example, Chen et al.⁵² used carbon felt as the cathode host material and applied inexpensive MnO₂ as the redox adsorbent for chlorine species to construct a Zn–Cl₂ battery. The mechanism of MnO₂ as the redox adsorbent is shown in Fig. 30a. During the charging process, Mn²⁺ is converted to MnO₂ and deposited onto the carbon felt cathode, while Cl⁻ is oxidized to Cl₂. Due to the adsorption of chlorine species by MnO₂, the produced Cl₂ is retained on the carbon felt, reducing its diffusion into the electrolyte. Additionally, a model for the interaction between Cl₂, MnO₂, and carbon felt was established, and the structural optimization, binding energy calculations, and charge density difference analysis were performed (Fig. 30b and c). The results indicate that Cl₂ adsorbs more easily onto MnO₂, and the introduction of MnO₂ leads to a rearrangement of the interfacial charge of chlorine, which accelerates the reduction of MnO₂ and facilitates the electrode reaction; this was further confirmed by XPS analysis (Fig. 30d). SEM results show that no solid material is generated on the cathode of a conventional Zn–Cl₂ battery (Fig. 30e–g). In contrast, for the Zn–Cl₂@MnO₂ battery, MnO₂ nanosheets are uniformly distributed on the cathode substrate, significantly enhancing the electrochemical performance of the aqueous Zn–Cl₂@MnO₂ battery. Additionally, Zhang et al.¹⁶⁹ employed carbonaceous cathode materials (graphene (G), AC, and N-doped AC (NAc)) as cathode host materials to further improve the energy density and operating output voltage of Zn–Cl₂ batteries. They used a high-concentration aqueous choline chloride (ChCl) electrolyte to expand the electrochemical stability window of the Zn–Cl₂ battery. The electrochemical performance of the 30 M ChCl electrolyte on various carbon-based substrates was evaluated. The results indicated that NAc exhibited a higher discharge capacity and the lowest internal resistance compared to G and Ac, suggesting that the battery with NAc as the substrate showed rapid charge transfer at the interface and performed more stable and faster charge/discharge cycles. Subsequently, the researchers performed FTIR (Fig. 30h) and XPS (Fig. 30i and j) analyses on the NAc host material to investigate the energy storage mechanism of the Zn–Cl₂ battery. The results showed that the battery mainly undergoes a reversible chloride ion insertion and extraction process during charging and discharging: during charging, Cl⁻ ions are inserted into the carbon-based material, while during discharging, Cl⁻ ions are extracted from the carbon structure. This mechanism reveals that the carbon material not only serves as an electron-conducting framework but also acts as a reversible ion carrier, participating in the energy storage reaction. This mechanism suggests that carbon materials in Zn–Cl₂ batteries function as electron transport channels and reversible chloride ion carriers, providing new insights into the development of efficient chloride ion storage systems.

Fig. 30 (a) Schematic of a Zn–Cl₂@MnO₂ battery and its working mechanism. (b) Adsorption modes and adsorption energies of chlorine on MnO₂ and carbon felt. (c) Charge density patterns of Cl-adsorbed MnO₂, MnO₂, and Cl₂, along with the corresponding charge density difference slices viewed from the top-down perspective. (d) Cl 2p XPS spectra of the cathodes after charging in the Zn–Cl₂ battery and after both charging and discharging in the Zn–Cl₂@MnO₂ battery. (e) SEM image of the conventional cathode of the Zn–Cl₂ battery after the second charge cycle. (f, g) SEM images of the cathode in the Zn–Cl₂@MnO₂ battery after the second charge and discharge cycles.⁵² Reproduced with permission from ref. 52. Copyright 2023, Wiley-VCH. (h) FTIR spectra of NAc carbon after charging and discharging. (i) Cl 2p and (j) C 1s XPS spectra of the NAc substrate at the fully charged state of the battery.¹⁶⁹ Reproduced with permission from ref. 169. Copyright 2024, Elsevier.

3.3.2. Functional frameworks. Functional frameworks, such as MOFs or COFs, provide a highly porous and tunable environment that can effectively accommodate halogen species during the charge and discharge processes. These structures can efficiently capture halogen molecules and prevent undesirable side reactions such as dissolution or shuttling,¹⁷⁰ thereby improving the electrochemical behavior of the battery. For example, Tulchinsky et al.¹⁷¹ designed and validated a MOF based on nitrogen-containing imidazole ligands. By the quantitative oxidation of halogen to coordinate the unsaturated Co(II) sites, a stable solid compound with terminal Co(III)-halogen bonds was generated (Fig. 31a). Starting from the precursor Co₂Cl₂BTDD (1), synthesized by reacting H₂BTDD with CoCl₂·6H₂O, Rietveld-refined neutron diffraction data showed that the structure consists of 1D helical –(Co–Cl)_n– chains and bis-imidazole bridge units, arranged in an ordered fashion along the honeycomb channels (Fig. 31b). In situ XAS studies revealed that this material can stably release halogens under heating conditions, and the conversion process almost completely reverses to the initial Co₂Cl₂BTDD (1) at approximately 285 °C, exhibiting good thermal responsiveness and structural reversibility (Fig. 31c). EXAFS analysis showed that, during thermal treatment, the scattering signal of the Cl shell gradually weakened, and the bond length of the first coordination shell elongated, indicating that most of the Co³⁺ in the MOF nodes was reduced to Co²⁺ (Fig. 31d and e). Although in situ XANES data indicated that the structural transformation was almost complete by the end of heating, thermogravimetric analysis (TGA) showed that the sample's mass loss corresponded to only about 70–80% of the halogen release, suggesting that some halogens remained confined in the structure (Fig. 31f). The discrepancy between XANES and TGA may arise from the fact that a small amount of Co³⁺ sites had not been fully converted, and these sites still contained tridentate Cl in their local coordination environment, leading to some structural distortion. It is speculated that the formation of X₂ is more likely through the homolytic cleavage of the Co³⁺–X bond, forming X˙ free radicals, which then couple to generate X₂. The incomplete transformation of Co³⁺ sites might restrict the migration and recombination of the free radicals due to the confinement environment or result in energy barriers due to structural microdifferences, further highlighting the crucial role of the host material in regulating the halogen release pathway (Fig. 31g and h). This study not only reveals the confinement and reversible adsorption regulation mechanisms of halogen species in MOF materials but also provides a feasible approach for constructing high-safety, controllable release, and high-capacity output Zn–Cl₂ cathodes. Especially in the context of the urgent need to address issues like electrolyte corrosion, frequent side reactions, and energy efficiency loss due to the high reactivity of chlorine gas, MOF materials can achieve stable storage and controlled release of active chlorine species due to their structural order, coordination responsiveness, and reversible thermal release properties, thereby effectively suppressing the shuttle effect, extending battery life, and improving system energy density. To more directly analyze the internal mesoscopic structure of MOF materials, the researchers used cryo-electron tomography (cryo-ET) technology to obtain high-resolution 3D imaging of their crystal morphology.¹⁷² The results showed that the MOF crystals exhibit an ordered cylindrical mesoporous structure with a radial distribution from the center outward, aligned perpendicular to specific crystal facets (Fig. 31i). Fast Fourier transform (FFT) images showed quadrupole symmetry along the [100] direction and dipole symmetry along the hexagonal mesoporous channel a-axis, further indicating that the material has a highly ordered structural feature at both the microscopic and mesoscopic scales (Fig. 31j). With the 3D high-resolution spatial distribution analysis achieved by cryo-ET, real-time monitoring of the structural evolution during the operation of zinc–halogen batteries can be achieved, further revealing the intrinsic correlation between the distribution, migration behavior, and confinement mechanisms of halogen species within the cathode host.

Fig. 31 (a) Schematic representation of the structure of the parent Co(II) MOF Co₂Cl₂BTDD and the secondary building unit structures of the Co centers in Co₂Cl₂BTDD, Co₂Cl₄BTDD, and Co₂Cl₂Br₂BTDD. (b) Powder XRD data of Co₂Cl₂BTDD, Co₂Cl₄BTDD, and Co₂Cl₂Br₂BTDD. (c) Theoretical XANES spectra of Co₂Cl₂BTDD and Co₂Cl₄BTDD, along with the room-temperature experimental XANES spectra of Co₂Cl₂BTDD, Co₂Cl₄BTDD, and thermally treated Co₂Cl₄BTDD. (d) Magnitudes and (e) imaginary components of the phase-uncorrected Fourier-transform k²-weighted EXAFS spectra of Co₂Cl₂BTDD, Co₂Cl₄BTDD, and thermally treated Co₂Cl₄BTDD. (f) In situ XANES spectra of Co₂Cl₂BTDD and Co₂Cl₄BTDD, along with linear combination analysis of the spectra across the entire series. (g) Possible mechanisms for the formation of elemental halogen, including the concerted, radical recombination, and radical attack. (h) Distances between terminal halogen atoms within the same pore in Co₂Cl₄BTDD and Co₂Cl₂Br₂BTDD.¹⁷¹ Reproduced with permission from ref. 171. Copyright 2017, American Chemical Society. (i) High-resolution image of meso-UiO-66 image using cryoET. (j) High-resolution cryo-TEM image of mesoporous UiO-66, along with the corresponding FFT pattern and the FFT-filtered TEM image extracted from the blue boxed region.¹⁷² Reproduced with permission from ref. 172. Copyright 2025, Springer Nature. DOS profiles of (k) Bi₂Se₃, (l) Cl@Bi₂Se₃, (m) Br@Bi₂Se₃, and (n) I@Bi₂Se₃ recorded under three different tip-approach currents at 1.0 V. (o) CB and VB energy of Bi₂Se₃, Cl@Bi₂Se₃, Br@Bi₂Se₃, and I@Bi₂Se₃. (p) Dirac point of Bi₂Se₃, Cl@Bi₂Se₃, Br@Bi₂Se₃, and I@Bi₂Se₃ recorded along the edge direction.¹⁷⁶ Reproduced with permission from ref. 176. Copyright 2018, Royal Society of Chemistry.

3.3.3. Other materials. The bulk phase of 2D topological insulators is insulating, while their surface states exhibit metallic properties, with a Dirac cone structure protected by time-reversal symmetry.¹⁷³ These materials demonstrate excellent characteristics such as high electron mobility, low scattering loss, strong robustness to defect disturbances, and tunable band structures.¹⁷⁴ By incorporating these topological surface states into electrode materials, efficient electron transport channels can be constructed, interface resistance can be reduced, and halogen shuttle effects can be suppressed,¹⁷⁵ significantly enhancing the rate performance and cycling stability of zinc–halogen batteries. However, there have been no reports in the literature on the direct application of topological insulator materials in zinc–halogen battery systems. Recent theoretical studies¹⁷⁶ have suggested that doping halogen elements (Cl, Br, and I) into the Se sites of Bi₂Se₃ can effectively modulate the energy gap between the Dirac point (ED) and the Fermi level (EF) (|ED-EF|), allowing precise tuning of the surface state band structure (Fig. 31k–n). Specifically, halogen doping with elements of different atomic numbers (Z) introduces charge carriers with varying valence, leading to an overall upward shift of EF. Additionally, the enhancement of spin–orbit coupling (SOC) induced by increasing Z further drives the Dirac point to shift inward within the band structure, thereby enhancing the reactivity of the surface states and the ability to catalytically confine species (Fig. 31o and p). Therefore, incorporating Bi₂Se₃-like topological materials into zinc–halogen systems, especially in Zn–Cl₂ batteries, would open new possibilities for the application of topological materials in zinc–halogen systems.

In the Zn–Cl₂ battery system, chlorine gas presents a significant challenge for the confinement and efficient conversion of chlorine species at the cathode due to its high reactivity and volatility. Carbon-based materials, with their excellent conductivity, rich porosity, and good chemical stability, have become ideal hosts for the adsorption, fixation, and reversible conversion of chlorine species, providing a solid structural foundation and electronic transport channels for constructing high-performance chlorine-based cathodes. In addition, when combined with redox mediators like manganese oxide, composite structures can further enhance the ability to capture Cl₂ and improve catalytic conversion efficiency, significantly boosting the battery's capacity output and cycling stability. Functional framework materials, due to their highly tunable structures and exceptional confinement ability, offer a novel paradigm for achieving stable adsorption, and structural reversibility control of chlorine gas. Furthermore, emerging studies based on 2D topological materials such as Bi₂Se₃ reveal their unique potential in constructing efficient electron transport interfaces and regulating surface reactivity, presenting cutting-edge possibilities for application in the Zn–Cl₂ system. Future research should further focus on fine-tuning the interfacial electronic structure, elucidating the synergistic mechanisms of confinement and catalysis, and combining in situ characterization techniques with theoretical simulations. This will enable the development of more targeted composite material systems, advancing the development of confinement–catalysis–conduction in constructing host materials, achieving precise regulation of chlorine species behavior, and comprehensively enhancing the energy density and cycling stability of Zn–Cl₂ batteries.

3.4. Halogen-confining host materials for Zn–dual halogen batteries

Zinc–halogen batteries exhibit good electrochemical performance based on halogen conversion reactions at the cathode and Zn²⁺/Zn deposition/stripping at the anode.³⁷ Compared to traditional zinc–halogen batteries, which are limited to two-electron conversion, the capacity and energy density of Zn–dual halogen batteries can be enhanced through unique strategies, such as activating the continuous redox reactions of halogen elements (i.e., dual halogen conversion chemistry), leading to the development of zinc–dual halogen batteries. To this end, Lv et al.²⁰ developed a quasi-solid-state zinc–dual halogen battery system with higher energy storage potential (Fig. 32a). This system consists of a self-supporting carbon cloth-iodine (CC-I₂) cathode and a high-concentration gel-state electrolyte containing iodide (I⁻) and bromide (Br⁻) ions, enabling a multi-step oxidation–reduction reaction of I⁻/I⁰/I⁺ and generating the key dual–halogen intermediate [IBr₂]⁻. Specifically, iodine is uniformly loaded onto the surface of the porous carbon cloth via a solution adsorption method, significantly improving its dispersion and reactivity. The introduction of LiBr not only serves as a source of Br⁻ to construct [IBr₂]⁻ but also increases the electrolyte concentration and reduces the free water content, thereby inhibiting side reactions on the zinc anode surface. The introduction of a small amount of LiNO₃ further stabilizes the zinc interface and suppresses uneven deposition. The confinement structure of the host carbon material plays a key role in the reaction process, where its vacancy defects and physical adsorption can effectively limit the migration behavior of halogen species, enhancing their controllability and utilization efficiency. Additionally, the carbon framework, rich in heteroatoms and defect sites, provides abundant active sites for halogen conversion reactions, exhibiting a synergistic confinement–catalysis effect. In situ Raman spectroscopy (Fig. 32b) shows the appearance of an I₅⁻ characteristic peak in the voltage range of 1.20–1.50 V, indicating the occurrence of the I⁻/I⁰ redox reaction. As the voltage rises to 1.50–1.67 V, this characteristic peak gradually shifts to the one corresponding to [IBr₂]⁻, revealing the continuous oxidation of iodine and the formation of the high-valent dual–halogen intermediate. Theoretical calculations (Fig. 32c) show that the Br–I–Br configuration has a negative Gibbs free energy, which is thermodynamically much more favorable than the unstable I–Br–Br configuration, confirming the spontaneity of this reaction pathway. Further charge analysis and potential distribution simulations (Fig. 32d) reveal strong charge interactions between halogens, which promote electron transfer and the occurrence of the redox process. Comparison experiments show that the system using a carbon cloth-Br⁻ cathode exhibits only a single platform at 1.75 V, with a capacity much lower than that of the CC-I₂ cathode system. CV curves further demonstrate that the involvement of iodine significantly improves the kinetics of the bromine oxidation reaction. Molecular orbital theory analysis reveals that the frontier orbital distribution of Br⁻ and [IBr₂]⁻ is conducive to rapid electron transfer (Fig. 32e). As expected, this quasi-solid-state zinc–dual halogen battery system shows excellent areal capacity and high capacity retention, demonstrating the significant potential of dual–halogen chemistry for aqueous energy storage systems. However, the complexity of operating in liquid-phase media to store high oxidation-state halogens still limits its widespread application in energy storage devices. To address this, Liu et al.¹⁷⁷ proposed a water-based zinc–dual halogen battery system that combines the scalability of flow batteries with the high specific capacity advantages of intercalation chemistry, introducing Br and Cl dual–halogen elements via molten hydrate electrolytes to form stable halozinc complexes as reversible active species, distinguishing it from traditional Zn–Br₂ battery systems. In this system, graphite acts as the key confinement host material, fixing halogen molecules via reversible intercalation reactions and providing ample active sites for reactions. XRD results show that during charging, graphite undergoes a multi-stage intercalation reaction, and after discharging to 1.0 V, the (002) crystal plane recovers, indicating the reversible intercalation and de-intercalation of halogen species while maintaining the integrity of the graphite structure (Fig. 32f). Raman spectroscopy analysis shows that the characteristic Raman stretching peaks of Br₂ and BrCl during the intercalation process are significantly lower than the intrinsic frequencies of free halogen molecules, reflecting significant charge transfer and electron coupling effects between the intercalated halogens and graphite, thereby enhancing their confinement-catalytic behavior (Fig. 32g). These Raman signals completely disappear after discharge, indicating the good reversibility of the halogen intercalation-de-intercalation behavior. Notably, the interlayer distance of the halogen intercalation species is significantly smaller than that of alkali metal ions or polyatomic anions, and the dense packing structure effectively reduces the interlayer electrostatic repulsion, allowing near-zero oxidation state halogens to exhibit higher stability in the graphite confinement structure. Electrochemical tests show that the charging overpotential of graphite increases gradually with the intercalation reaction, while the discharge process has a lower overpotential, indicating that the kinetics of the intercalation process are more limiting than the de-intercalation process. GITT testing reveals that the halogen diffusion coefficient is much higher than that of Li⁺ in LiFePO₄ and is close to the recently reported diffusion performance of PF₆⁻ in graphite, indicating that graphite provides efficient diffusion channels and reaction activity for halogens (Fig. 32h and i). This battery system maintains a stable capacity of approximately 90 mAh g⁻¹ after 100 cycles at 0.25 A g⁻¹ (Fig. 32j), thanks to the intercalation accommodation capacity of graphite for halogens and the ability of the molten hydrate electrolyte to mitigate the diffusion and self-discharge issues of polybromide ions. From a material selection perspective, in addition to graphite, new materials with adjustable structures and confinement channel features, such as MOFs and 2D transition metal sulfides, are also considered potential halogen confinement and catalytic materials, providing more possibilities for the performance improvement and structural optimization of aqueous zinc–dual halogen batteries.

Fig. 32 (a) Development of quasi-solid-state zinc dual–halogen batteries. (b) Voltage profiles of the zinc dual–halogen battery, along with the in situ Raman spectra of the carbon cloth iodine cathode and fitted Raman spectra at selected voltages. (c) Theoretical calculations of the free energy differences between Br⁻ and IBr with [IBr₂]⁻. (d) Atomic charge calculations of Br and I in [IBr₂]⁻. (e) Energy gaps, LUMOs and HOMOs of [IBr₂]⁻ on the electrode surface and Br⁻ in the electrolyte.²⁰ Reproduced with permission from ref. 20. Copyright 2022, American Chemical Society. (f) GCD curves of the graphite electrode, along with XRD patterns and (g) Raman spectra collected at corresponding voltages. (h) GITT charge–discharge curves and (i) halogen diffusion coefficient (D_x) of the graphite cathode. (j) Cycling performance and GCD curves of a zinc dual–halogen battery with ZnCl₂·0.03KBr·2H₂O electrolyte at 0.25 A g⁻¹.¹⁷⁷ Reproduced with permission from ref. 177. Copyright 2020, Wiley-VCH.

In summary, zinc–dual halogen batteries have effectively overcome the energy density limitations of traditional single-halogen systems by introducing multi-electron redox pathways, constructing key dual–halogen intermediates and designing carbon-based hosts or graphite intercalation structures with confinement and catalytic functions. Notably, innovations in quasi-solid-state electrolyte systems and molten hydrate media have significantly improved the fixation and conversion efficiency of halogen species, exhibiting excellent cycling stability and rate performance. However, this system still faces numerous challenges in terms of long-cycle stability and the conversion efficiency of interface controllability. Future research can focus on the following directions: (i) further designing structurally tunable and interface-controllable multidimensional confinement materials (such as MOFs, 2D layered materials, etc.) to achieve efficient capture and reversible conversion of halogen species, expanding new pathways for constructing confinement–catalysis–conduction host materials; (ii) constructing new dual–halogen systems to explore more controllable multi-electron intermediates; (iii) developing electrolytes with high ion conductivity and good interface stability, unlocking the potential for higher energy density and longer cycling stability.

To contextualize the development status of zinc–halogen batteries within the broader landscape of metal-ion technologies, we present the performance parameters of zinc–halogen batteries in pouch or coil cell configurations. As summarized in Table 1,^{19,20,52,86,160,178–192} zinc–halogen batteries typically exhibit a relatively low operating voltage, satisfactory specific capacity, modest energy density, and acceptable cycling stability. While their cost-effectiveness and operational safety are prioritized, these characteristics make zinc–halogen systems particularly suitable for stationary and grid-scale applications. In contrast, emerging Li-, Na-, and K-ion batteries generally offer higher energy densities but suffer from persistent issues such as thermal instability and dendrite formation, especially under high-rate or prolonged cycling conditions^193,194 The superior ion transport kinetics of zinc–halogen systems, relative to Mg-ion and Al-ion batteries, can be attributed to the higher mobility and lower hydration enthalpy of Zn²⁺ in aqueous electrolytes.¹⁹⁵ Furthermore, zinc–halogen batteries typically employ mild, non-corrosive aqueous electrolytes that are safer and more environmentally friendly,¹⁹⁶ in contrast to the often corrosive and complex electrolytes required for Mg-ion and Al-ion batteries. These combined advantages contribute to lower electrode polarization and higher energy efficiency, as the electroplating and stripping of Zn are more reversible and exhibit lower overpotentials than those of Mg or Al. Nevertheless, it should be noted that the energy density and cycling stability of zinc–halogen batteries still require significant improvement. Therefore, while their current advantages define their position in the actual state-of-the-art of emerging systems, further innovations are necessary to fully unlock their potential for next-generation applications.

Table 1 Comparative performance parameters of zinc–halogen batteries based on recent literature reports

Types of zinc–halogen batteries	Electrolyte types	Specific capacity (mAh g^−1)/Current density (A g^−1)	Operating voltage (V)	Maximum energy density (Wh kg^−1)	Coulombic efficiency (%)	Capacity retention/cycles numbers/current density (A g^−1)	Ref.
Zn||Cl2@MnO2	5 mM MnSO4 + 1 M ZnSO4 + 1 M LiCl	—	1.4–2.5	—	91.6	94.7%/1000/2.5 mA cm^−2	52
Zn||I2@MBene-Br	21 M LiTFSI + 1 M Zn(OTf)2	143.8/1	0.4–2.1	485.8	—	—	180
Zn||tetrapropylammonium bromide	1 M Zn(DFTFSI)2	288.9/0.1	1–1.8	—	99.8	92.5%/1200/2	181
Zn||N-hexylpyridinium bromide	3 M ZnSO4	217/1C	0.4–1.95	106	99.8	88.5%/1000/1C	182
Zn||gel||carbon cloth-I2	Concentrated aqueous gel electrolytes	635/3 mA cm^−2	0.6–1.8	3.11 mWh cm^−2	—	82%/500/3 mA cm^−2	20
Zn||Br2	20 M Li[NTf2] + 3 M Zn[OTf]2	—	1–2	15.89 Wh L^−1	—	76.8%/1000/0.5 mA cm^−2	178
Zn||Br2@FeSAC-CMK	2 M ZnSO4	270/2	1–1.85	585	99	88%/2000/2	183
Zn||TBABr3-solids@C composite	Ternary-hydrated eutectic electrolyte	283.4/0.5	0.5–1.85	292	96.5 (pouch cells)	85%/3000/0.5	184
Zn||KBr-Oc4NBr	3 M ZnSO4	206/3C	1–1.85	/	99.92	98.46%/775/3C	19
Zn||Sb/Sb2Zn3	0.5 M ZnBr2 + 0.25 M TPABr	—	0.5–1.95	62	98.5	Almost 100%/800/10 mA cm^−2	185
Zn||TI2CTx MXene	ZnBr2	193 μAh cm^−2/1.0 mA cm^−2	0.5–1.9	—	85	87%/3000/1 mA cm^−2	160
Zn||MXene-AB@I	2 M ZnSO4 + 0.1 M Li2SO4	188.3/0.1	0.6–1.6	171.3	99.14	95.1%/260/0.1	86
Zn||I2@active carbon	0.4 wt% Borax-bacterial cellulose/p(AM-co-VBIMBr) + 2 M Zn(ClO4)2	550/1	0.50–1.80	688	97.5	81.0%/2000/1	186
Zn||I2	1 M ZnSO4 + 0.5 M trimethylamine hydrochloride	450/2	0.60–1.90	—	—	70.0%/5000/2	187
Zn||Fe–N–C encapsulated carbon	1 M ZnSO4 + 9 M KI	460/0.5	0.40–1.60	253	99	84%/100/3 mA cm^−2	179
Zn||I2	Dimethyl sulfone and niacinamide + Zn(ClO4)2·6H2O	412/0.5	0.50–1.90	404	98.9	80%/2000/2	188
Zn||N,F-codoped porous carbon	2 M ZnSO4+ 0.1 M ZnI2 + 0.1 M ZnBr2	440.3/1	0.60–1.80	584.1	—	89.6%/8000/5	192
Zn@Cu||I2@KB2	2 M ZnSO4 + 0.1 M BMIS	6.5 mA h cm^−2/2.4C	0.6–1.6	97.34	99.98	88.3%/5000/0.4C	189
Zn||I2/PPCMK	2 M ZnSO4	236.76/1C	0.5–1.6	290.50	99.73	95.08%/20000/20C	190
Zn||single-atom Cu-embedded N-doped Ketjen black	0.5 M ZnSO4 + 0.5 M Li2SO4	121/5	0.5–1.6	—	98.4	92.5%/5000/5	191

---

4. Optimization strategies of halogen-confining host materials for zinc–halogen batteries

To address challenges associated with cathodes in zinc–halogen batteries, researchers have proposed a series of optimization strategies to modify various types of halogen-confining host materials to enhance their electrochemical performance. These strategies include rational structural design, surface functionalization, heteroatom doping, atomic-level engineering of SACs/DACs, and heterostructure engineering. In the following sections, we discuss the specific applications and recent advancements in these modification methods based on representative studies.

4.1. Rational structural design

Carbonaceous materials are the most widely used halogen-confining host materials in zinc–halogen batteries due to their porous structures, which can accommodate halogen species. However, they exhibit suboptimal anchoring capabilities for halogen species. The structural design of carbonaceous materials has proven to be an effective approach for improving their anchoring performance.¹⁹⁷ From the perspective of physical confinement, the pore size and structure of carbonaceous materials play a critical role in spatially constraining halogen species.¹⁹⁸ Existing studies have demonstrated that adjusting the nanopore structures of carbonaceous materials (e.g., micropores, mesopores, and macropores) can significantly enhance physical interactions between halogen species and host materials.¹⁹⁹ Micropores (<2 nm) are advantageous for the adsorption of halogen species due to their size compatibility. However, the capillary effect within micropores can lead to the formation of numerous “dead” pores in the absence of sufficient electrolyte infiltration, thus reducing the utilization of active sites for halogens.²⁰⁰ In contrast, mesopores (2–50 nm) facilitate ion diffusion, allowing the electrolyte to penetrate the pores more thoroughly and rapidly, thus enhancing capillary action.²⁰¹ Consequently, constructing hierarchical porous structures that integrate micropores and mesopores has emerged as a key strategy for optimizing carbonaceous materials. The hierarchical porous structures not only anchor halogen species firmly within the micropores but also accelerate ion transport via mesopores, providing highly efficient pathways for redox reactions and significantly enhancing the electrochemical performance of zinc–halogen batteries. Wang et al.²⁰² designed and successfully prepared bimodal highly ordered mesoporous carbon (BOMC) for Zn–Br₂ batteries (Fig. 33a). Both the specific surface area and adsorption performance were significantly enhanced by controlling the morphology of BOMCs (Fig. 33b–g) and introducing 2 nm pores within the walls of 5 nm mesopores. This structural optimization provided abundant active sites for the Br₂/Br⁻ redox reaction, facilitating bromine adsorption. The highly ordered mesostructure effectively improved the Br₂ diffusion coefficient by shortening the mass transfer pathway, increasing the mass transfer rate, and reducing diffusion resistance. As a result, Zn–Br₂ batteries based on the optimized BOMC-2 exhibited 82.9% voltage efficiency and 80.1% energy efficiency at 80 mA cm⁻², demonstrating excellent cycling stability. However, it remains challenging to construct carbon-based halogen host materials that simultaneously exhibit excellent electrical conductivity and well-designed pore structures in a simple and cost-effective manner. To this end, Wu et al.²⁰³ synthesized a carbon host (PTCC900) by annealing perylene-3,4,9,10-tetracarboxylic dianhydride (PTCDA) at 900 °C, which served as an iodine host to form the PTCC900@I₂ cathode for Zn–I₂ batteries (Fig. 33h). PTCC900 features a hierarchical porous structure with pore sizes ranging from 1.8 to 13 nm, effectively confining I₂ and polyiodides. Moreover, the turbostratic graphitic microstructures inside PTCC900 create fast electron transport pathways, significantly enhancing electrical conductivity. Consequently, the PTCC900@I₂ cathode delivered high capacity, outstanding cycling stability, and remarkable rate performance. To develop halogen species host materials with stronger adsorption capabilities and ordered pore structures, Xu et al.²⁰⁴ developed host materials by using Ca²⁺ and Zn²⁺ to adjust the structure and synthesized hierarchical micro-mesoporous carbon nanospheres (HMMC NSs) with uniform pore distribution (Fig. 33i). The unique hierarchical micro-mesoporous structure enhances the material's ability to immobilize nanoscale iodine molecules, effectively suppressing the shuttle effect. Simultaneously, the ordered pore architecture ensures uniform iodine loading, significantly reducing iodine agglomeration during the reaction process. The synergistic regulation of the hierarchical micro-mesopores and the ordered pore arrangement enables Zn–I₂ batteries to exhibit high reaction kinetics, as evidenced by the CV curves of the HMMC-I₂ NS cathode at different scan rates (Fig. 33j). Notably, the b-values of the oxidation and reduction peaks are both close to 1, indicating that the charge storage process is predominantly governed by capacitive behavior (Fig. 33k). Moreover, the capacitive contribution of the HMMC-I₂ NS cathode remains dominant across various scan rates (Fig. 33l), further confirming its excellent electrochemical reaction kinetics. These studies demonstrate that structurally optimized carbon-based materials, particularly those with hierarchical pore structures and excellent conductivity, not only efficiently confine halogen species and suppress the shuttle effect but also significantly enhance ion/electron transport efficiency and redox kinetics. This highlights the critical role of structural design of host materials in enabling the confinement–catalysis–conduction paradigm for high-performance zinc–halogen batteries.

Fig. 33 (a) Synthesis process and mechanism of action of bimodal ordered mesoporous carbon. SEM images of (b) and (c) BOMC-0, (d) and (e) BOMC-1.²⁰² Reproduced with permission from ref. 202. Copyright 2016, Elsevier. SEM images of (f) PTCC900 and (g) PTCC900@I₂. (h) Fabrication process of PTCC900@I₂.²⁰³ Reproduced with permission from ref. 203. Copyright 2023, Elsevier. (i) Schematic of the synthesis process of HMMC-I₂. (j) CV curves of HMMC-I₂ NS recorded at different scan rates. (k) log(i)–log(v) plots of the oxidation and reduction peak currents of the HMMC-I₂ NS cathode. (l) Capacitive contribution and diffusion-controlled contribution versus the scan rate of the HMMC-I₂ NS cathode.²⁰⁴ Reproduced with permission from ref. 204. Copyright 2023, Elsevier.

Despite recent progress in Zn–Br₂ batteries, several key challenges remain, including slow bromine redox reaction kinetics, uncontrollable diffusion of active bromine, and the easy growth of zinc dendrites.²⁰⁵ To overcome these bottlenecks, it is crucial to focus on material structural design, creating multidimensional systems with precise spatial confinement and ion channel regulation capabilities. To this end, Li et al.²⁰⁶ proposed hollow core–shell carbon spheres (HCSC) as functional electrode materials based on the structural confinement design concept (Fig. 34a). TEM analysis revealed that the HCSC exhibited a typical core–shell structure, with a core diameter of approximately 278 nm and a shell diameter of approximately 404 nm (Fig. 34b). This structure not only provides an ideal confinement space for bromine but also offers convenient channels for electrolyte infiltration and ion diffusion. Thanks to the designability of the core–shell structure, HCSC combines high specific surface area, rich pore channels, and excellent conductivity, significantly enhancing the bromine fixation capability in the cathode region while effectively suppressing its diffusion to the anode, thus significantly alleviating the battery's self-discharge behavior. Test results show that after 24 hours of rest at a certain current density, the capacity retention of the battery with the HCSC-CF cathode was significantly better than other comparative electrodes, demonstrating the prominent effect of HCSC in enhancing bromine fixation and suppressing self-discharge (Fig. 34c). Furthermore, the battery maintained low polarization and stable performance during cycling at high current densities, further validating the excellent performance of HCSC in regulating bromine behavior and stabilizing the interface reaction (Fig. 34d). In addition to regulating bromine behavior, HCSC also showed a positive effect on zinc deposition behavior. Multi-physics simulation results indicated that, compared to conventional electrodes, the Zn²⁺ concentration and electric field distribution on the HCSC-modified electrode surface were more uniform, facilitating uniform nucleation and deposition of zinc (Fig. 34e and f). Specifically, the rich defect structure and hierarchical pore channels on the HCSC surface provide abundant reaction sites for zinc deposition and induce ordered deposition within its cavity structure, effectively suppressing the disordered growth of dendrites (Fig. 34g), thereby significantly improving cycling stability and battery lifespan. This study reveals the application potential of hollow-core carbon spheres as host materials under the confinement–catalysis–conduction synergy, providing a new design approach for high-performance zinc–halogen batteries. To further achieve efficient confinement and stable redox regulation of iodine species, Zhang et al.²⁰⁷ proposed a host material design strategy for non-metallic sodium/lithium-based systems, using heteroatom-doped porous graphite carbon (HPCM-NP) as a support material. This material, based on a cellulose wiping cloth, was synthesized with a polyaniline coating layer through the aniline interface polymerization assisted by phytic acid, followed by carbonization, forming HPCM-NP materials with a multi-level pore structure and excellent conductivity (Fig. 34h). They employed in situ Raman spectroscopy to track the reaction path of iodine during the charge and discharge process (Fig. 34i). Results showed that the characteristic I₃⁻ peak appearing at the beginning of discharge indicated that iodine first transformed into I₃⁻ and cooperatively intercalated with sodium ions. As the peak gradually disappeared, corresponding to the conversion from NaI₃ to NaI, the reversible I₂ → I₃⁻ → I⁻ reaction process was confirmed. Performance tests showed that the iodine–carbon composite cathode material, in a full battery system without a Zn anode, exhibited high reversible capacity under various electrolyte conditions. This design strategy combines the surface redox characteristics of iodine with the ion intercalation advantages of porous carbon, effectively avoiding the safety risks posed by the Zn anode, providing a new perspective for the development of non-metallic batteries based on iodine systems. To summarize, constructing host materials with multi-level pore channels and controllable morphologies through structural design can provide rich loading and confinement sites for halogen species. Particularly, by regulating pore size, pore distribution, and structural order, it helps enhance the physical adsorption and interface coupling between halogens and carbon materials, thereby improving their confinement ability and reaction controllability.

Fig. 34 (a) Structure of the HCSC and the associated mechanisms of zinc deposition regulation, catalysis and bromine immobilization. (b) TEM images of HCSC. (c) Capacity retention of Zn–Br₂ flow batteries (ZBFBs) assembled with pristine CF (PCF), CS-modified CF (CS-CF), and HCSC-modified CF (HCSC-CF) cathodes at 80 mA cm⁻². (d) Charge–discharge curves of the ZBFB with the HCSC-CF cathode at 160 mA cm⁻². Schematic diagrams of (e) Zn²⁺ concentration and (f) electric field distribution on HCSC-CF and PCF electrodes, respectively. (g) Zinc deposition mechanisms on HCSC-CF and PCF electrodes.²⁰⁶ Reproduced with permission from ref. 206. Copyright 2025, Wiley-VCH. (h) Preparation process of HPMC-NP and the mechanism of iodine encapsulation. (i) Charge–discharge curves of the Na–I₂ battery with I₂-HPMC-NP as the cathode, along with the corresponding in situ Raman spectra at various charge–discharge voltages.²⁰⁷ Reproduced with permission from ref. 207. Copyright 2017, Springer Nature.

Despite these advancements, current structural design strategies still face several key challenges: (i) while enhancing the anchoring ability and adsorption performance of carbon materials for halogen species, there is a lack of systematic optimization studies that also consider their electrical conductivity and mechanical structural stability; (ii) the regulation of pore structures in host materials remains rudimentary, with insufficient coordination between pore size, distribution uniformity, and spatial connectivity, making it difficult to simultaneously meet performance demands such as electrolyte penetration, rapid ion migration, and full exposure of active sites; (iii) most carbon host materials still primarily rely on physical confinement mechanisms and lack effective control over the electronic structure and reaction pathways of halogen species, limiting the enhancement of halogen redox reaction kinetics. To overcome these bottlenecks, future research should focus on constructing multidimensional synergistic structural systems, such as introducing electron-rich regions via heteroatom doping to enhance chemical anchoring, optimizing ion distribution and charge transfer processes through interface polarity regulation, or constructing ordered hierarchical pores to improve mass transfer efficiency and interface reactivity. These strategies are expected to enable the synergistic coupling of confinement adsorption, electrocatalytic conversion, and ion conduction, providing a structural paradigm for the rational design of efficient cathode materials for zinc–halogen batteries.

4.2. Surface functionalization

The surface functionalization of carbonaceous materials can effectively enhance the strong chemical interaction between the host material and halogen species by increasing the polarity of the carbonaceous material.²⁰⁸ Furthermore, it can improve the hydrophilicity of the material, facilitating better contact with the electrolyte and enhancing the diffusion of reactants and the efficiency of interfacial reactions. Jin et al.²⁰⁹ fabricated CNTs modified with oxygen-containing functional groups (HCNT-O) via an in situ electro-deposition strategy (Fig. 35a). SEM images revealed that the HCNT-O material maintained the typical carbon nanotube morphology (Fig. 35b). In situ Raman testing further revealed its good electrochemical reversibility: as the charging voltage increased to 1.5 V, the characteristic peaks of I₃⁻ (108 cm⁻¹ and 223 cm⁻¹) gradually intensified, and they quickly disappeared when discharged to 0.2 V (Fig. 35c), indicating the high reversibility of the I₂/I₃⁻ conversion process. The HCNT-O structure not only provides a high-conductivity electron transport framework for the electrode but also effectively confines I₃⁻ through its pore structure, thereby suppressing the shuttle effect of iodine species. In addition, this structure helps guide uniform zinc (Zn²⁺) reduction and deposition, effectively suppressing dendrite formation, significantly improving the cycling stability and capacity retention of the battery. The full-cell test achieved a volumetric energy density of 1647.3 mWh cm⁻³, outperforming most high-performance Zn/Li-ion micro-supercapacitors. This study fully demonstrates the key role of surface-functionalized carbon nanotube structures in enhancing iodine reaction reversibility, suppressing active species migration, and increasing device energy density. Given the significant role of oxygen-containing functional groups in improving the iodine species confinement performance of carbon hosts, subsequent research further introduced nitrogen doping strategies. By constructing porous nitrogen-doped carbon structures, they achieved synergistic regulation of charge distribution and surface polarity to enhance the adsorption and electrocatalytic conversion efficiency of active species. For example, Wang et al.²¹⁰ constructed a 3D porous nitrogen-doped graphene material with a wave-folded structure, and the uniformly distributed oxygen-containing functional groups on the surface provided excellent polarity and interfacial reactivity (Fig. 35d). This material, by adjusting the surface properties of the cathode, formed a synergistic effect with high-concentration electrolytes, effectively stabilizing the electrode/electrolyte interface and improving the CE and capacity retention of the battery (Fig. 35e). To explore the redox mechanism of iodine in high-concentration electrolytes, they used in situ Raman spectroscopy to track the reaction process in fiber batteries. The results showed that the characteristic peaks of I₃⁻ intensified during charging and weakened during discharge, while the ICl peak appeared during charging and disappeared after discharge. Simultaneously, the signal of [Zn(OH₂)₂Cl₄]²⁻ changed synchronously with I₃⁻, indicating that it migrated to the cathode during charging, releasing Cl⁻, which helped stabilize I⁺ and form ICl (Fig. 35f). DFT calculations confirmed that graphene containing oxygen functional groups such as C(O)OH, C=O, and C–OH exhibited significantly lower adsorption energy for I₃⁻ than pure graphene, indicating that these polar functional groups significantly enhance the chemical confinement ability for iodine intermediate species (Fig. 35g). Building on this, Qu et al.²¹¹ proposed a “two birds with one stone” strategy combining directional regulation with adsorption stability by introducing sulfonate groups to modify carbon fibers, achieving synergistic regulation of iodine species confinement at the cathode and zinc deposition behavior at the anode. UV-vis absorption spectroscopy (Fig. 35h) further confirmed that the modified SO₃H-150 had a stronger anchoring ability for iodine species. DFT calculations revealed that sulfonate modification significantly reduced the adsorption energy of iodine species, thereby enhancing its confinement ability (Fig. 35i). Additionally, differential charge density analysis showed that the sulfonate group exhibited strong electron-withdrawing properties, significantly modulating the charge distribution on the carbon substrate surface, enhancing electron coupling and interactions with iodine species and further improving adsorption stability. These studies not only reveal the coupling mechanism of surface functionalization in regulating the stability, charge transfer behavior, and interfacial reaction kinetics of iodine species but also expand the new paradigm for constructing confinement–catalysis–conduction host materials.

Fig. 35 (a) Preparation process of the Zn–I₂ microbattery (ZIDMB) electrode. (b) SEM image of HCNT-O. (c) Charge–discharge curves of ZIDMBs and the corresponding in situ Raman spectra during the electrochemical process.²⁰⁹ Reproduced with permission from ref. 209. Copyright 2022, Wiley-VCH. (d) SEM and EDS images of GF-180. (e) Synergistic optimization mechanism between electrode design and electrolyte regulation. (f) Charge–discharge curves of GF-180 at 1 A cm⁻³ and the corresponding in situ Raman spectra during the electrochemical process. (g) Adsorption energy of I₃⁻ on pristine graphene sheets and graphene functionalized with C–OH, C=O, and C(O)OH groups.²¹⁰ Reproduced with permission from ref. 210. Copyright 2024, Royal Society of Chemistry. (h) UV-vis spectra of Zn(I₃)₂ solutions after 48 h of immersion with NCNF and SO₃H-150, respectively. (i) Strategy employing sulfonate-functionalized carbon fibers to suppress iodine dissolution and shuttling and to promote uniform Zn²⁺ deposition.²¹¹ Reproduced with permission from ref. 211. Copyright 2024, Science China Press.

Recently, researchers have introduced sulfonate groups into ordered framework material systems, such as COFs, to achieve distinct interface regulation performance and structural tuning capabilities. For example, Huang et al.²¹² designed a sulfonate-functionalized covalent organic framework (DMSBA-Tp-COF) and combined it with glass fiber (GF) and graphene (Gr) to construct a composite membrane Gr@DMSBA-Tp-COF@GF (Fig. 36a) for regulating the interface reactions in iodine-based batteries. The COF material features a 2D layered structure and abundant 1D channels, with the introduced –SO₃⁻ functional groups facilitating rapid Zn²⁺ migration and inducing uniform zinc deposition (Fig. 36b). At a current density of 20 mA cm⁻², the assembled battery achieved a high areal capacity of 3.2 mAh cm⁻² and demonstrated good cycling stability, highlighting the significant advantages of this confined COF membrane in enhancing energy density and interface stability while providing ideal confinement for iodine species. Compared to post-functionalized COFs, the ligand functionalization of MOFs can introduce electroactive sites and polar functional groups during the early stages of structural construction, potentially further improving confinement adsorption, charge distribution regulation, and reaction path induction. Recently, Zhang et al.²¹³ selected terephthalic acid derivatives with substituents such as –H, –OH, –(OH)₂, –NH₂, –F₄, –Cl₄, –Br, and –Br₂ as ligands and coordinated them with vanadium metal ions to form structurally diverse X-MIL-47 materials by using ML, which were further assembled into cathodes (Fig. 36c). DFT calculations showed that the different polar substituents significantly impacted LUMO energy levels of the ligands based on terephthalic acid (TPA), with the decreasing order of LUMO energy being F₄-TPA < Br-TPA < (OH)₂-TPA < H-TPA < OH-TPA < Br₂-TPA < NH₂-TPA < Cl₄-TPA (Fig. 36d). Substituents such as F₄ and Br notably enhance the reduction potential of the ligands, improving their redox activity. Additionally, Br-TPA has a narrower bandgap, indicating its stronger electron conductivity. MESP analysis (Fig. 36e) revealed that, compared to H-TPA, ligands containing –OH, –(OH)₂, –NH₂, –F₄, –Br, and –Br₂ exhibited higher electron density around the carboxyl oxygen atoms, reducing the C=O bond energy, making it easier to acquire electrons during reduction and enhancing coordination with vanadium sources. XAS further confirmed the enhancement of local electronic density around the V center due to polar substituents (Fig. 36f and g). EXAFS analysis showed a significant V–O coordination scattering peak at an R space of 1.59 Å (Fig. 36h and i), indicating a clear V–O coordination environment and high structural stability. To explore the regulation mechanism of functionalized MIL-47 for Zn²⁺ storage, the authors compared the VBr-160 and VBr-180 electrodes under different heat treatment conditions. Results showed that the XRD peaks of VBr-160 hardly changed during charging and discharging, indicating that its lattice is insensitive to Zn²⁺ insertion, with weak zinc storage capacity; in contrast, the (020) and (022) crystal facets of VBr-180 shifted to lower angles during discharge and returned during charging, reflecting good reversible Zn²⁺ insertion/extraction behavior (Fig. 36j). Raman testing further confirmed the reversibility of this process (Fig. 36k). These studies further demonstrate that the introduction of polar substituents can induce charge redistribution within the MOF system, enhancing the reversibility of the redox process and the migration kinetics of Zn²⁺. This work innovatively incorporates AI methods into the selection and structural design of functionalized ligands, enabling efficient prediction and regulation of molecular orbital energy levels, charge distribution, and interfacial electronic coupling. This approach provides a new design paradigm for developing advanced host materials with enhanced confinement capability and reversible conversion characteristics.

Fig. 36 (a) Fabrication process of Gr@DMSBA-Tp-COF@GF. (b) Cycling performance of Zn||Cu half-cells with GF, Gr@GF, and Gr@DMSBA-TP-COF@GF separators at 5 mA cm⁻².²¹² Reproduced with permission from ref. 212. Copyright 2025, KeAi Publishing. (c) Schematic diagram of the structure of MIL-47(V), TPA ligands with various polar groups, and the application of modified MIL-47(V) in zinc-ion batteries. (d) LUMO and HOMO energy levels and energy gaps of TPA molecules containing different polar functional groups. (e) MESP plots of TPA with different functional groups. (f) and (g) XANES spectra of X-MIL-47 functionalized with electron-donating and electron-withdrawing polar functional groups, along with the corresponding (h) and (i) Fourier-transform spectra. (j) Charge–discharge curves of the VBr-180 cathode with the corresponding XRD patterns and (k) Raman spectra at the respective states.²¹³ Reproduced with permission from ref. 213. Copyright 2025, Wiley-VCH.

In summary, this section discusses the key role of surface-functionalized halogen host materials in enhancing the performance of zinc–halogen batteries, highlighting the significant advantages of incorporating oxygen-containing or sulfonate functional groups in improving halogen species anchoring, suppressing shuttle effects, regulating electrode interface behavior, and promoting uniform zinc deposition. However, current surface functionalization strategies still face several challenges, such as difficulties in precisely controlling the type of functionalization, the extent of functionalization, and the spatial distribution of functional groups, which can lead to uneven interfacial reactions and hinder electron transport. Some functional groups also exhibit poor stability during cycling and may degrade in the electrolyte, affecting battery lifespan. Additionally, materials like COFs/MOFs have complex synthesis processes and limited conductivity, restricting their use in high-power energy storage applications. Thus, constructing integrated design strategies that combine confinement and guiding functions to regulate halogen species fixation at the cathode and zinc deposition at the anode promotes the development of efficient and synergistic zinc–halogen battery systems.

4.3. Heteroatom doping

Heteroatom doping strategies have attracted considerable attention among various methods for modifying halogen-confining host materials due to their simplicity and high efficiency. By doping heteroatoms (such as nitrogen, phosphorus, or sulfur) into carbonaceous materials, the surface charge environment of adjacent carbon atoms can be adjusted, thereby creating polar surfaces and increasing the number of active sites.²¹⁴ Further, this modification enables a transition from hydrophobic to hydrophilic behavior or from chemical inertness to high chemical activity, which enhances the chemical adsorption of halogen species.²¹⁵ Nitrogen atoms, due to their similar atomic radius and higher electronegativity, can readily substitute for carbon atoms. As a result, the nitrogen-doping strategy has been widely explored. To further understand the intrinsic catalytic activity of porous carbons doped with different N-heteroatoms for the iodine redox reaction and their correlation with battery performance in Zn–I₂ batteries, Liu et al.²¹⁶ reported a nitrogen-doped porous carbon material (PNC) constructed using ZIF-8 as the precursor. ZIF-8 is synthesized by co-precipitating a nitrogen-rich methyl imidazole ligand with zinc nitrate, followed by pyrolysis to form nitrogen-doped carbon with different structures, used as the key host material to inhibit iodine intermediate formation (Fig. 37a). Computational studies showed that graphite nitrogen significantly lowers the dissociation energy barrier of the I₃⁻ intermediate, facilitating its further reduction to I⁻, while also weakening the energy barrier required for the nucleation and deposition of solid iodine, thereby increasing the utilization efficiency of the active species. Such graphite nitrogen-rich PNC-1000 materials provide Zn–I₂ batteries with excellent rate performance and ultra-long cycling life. Based on transition state search calculations (Fig. 37b), the dissociation behavior of I₃⁻ is highly sensitive to the nitrogen source configuration. Among them, graphite-type nitrogen, due to its excellent charge delocalization and electronic abundance, significantly reduces the energy barrier of the reaction transition state, promoting the conversion of I₃⁻ to I⁻, demonstrating outstanding electrocatalytic performance. Compared to pyrrole nitrogen, pyridine nitrogen, or undoped carbon frameworks, graphite nitrogen is more beneficial for stabilizing I₃⁻ adsorption and accelerating its dissociation, opening up the two-electron conversion path for I₂/I⁻ and effectively avoiding side reactions caused by intermediate accumulation (Fig. 37c). UV titration results (Fig. 37d) further confirmed the critical role of graphite nitrogen in identifying and regulating the I₃⁻ intermediate, demonstrating its efficient capture and conversion ability for halogen species, which provides an important basis for host material design, contributing to the high and stable operation of zinc–halogen batteries. Previous studies have confirmed that the strong interaction between N dopants and iodine can assist iodine nucleation by lowering the activation energy and overpotential during the charging process.³⁹ Compared to single-atom doping strategies, Ye et al.²¹⁷ constructed a concave polyhedral structure of NZ-aNC (Ni/Zn bimetallic ZIF-derived activated nitrogen-doped carbon with air pre-activation) carbon hosts through Ni/Zn synergistic guidance and nitrogen source self-doping. This structure, with abundant defects and strong adsorption sites, effectively inhibits Br⁻ migration, accelerates redox kinetics, and significantly enhances interface stability and active material utilization. In situ UV-vis spectra (Fig. 37e and f) indicated that the NZ-NC (Ni/Zn bimetallic ZIF-derived nitrogen-doped carbon without the air pre-activation)@Br₂ electrode exhibited Br₃⁻ absorption peaks during charging and discharging, indicating limited fixation of halogen species and more intermediates in the reaction path. In contrast, the NZ-aNC@Br₂ electrode only detected Br⁻ characteristic signals, suggesting its ability to catalyze the direct conversion of Br₂ to Br⁻ and effectively suppress Br₃⁻ generation and accumulation. Raman spectra (Fig. 37g and h) revealed a prominent Br₅⁻ stretching peak in the case of the NZ-NC@Br₂ electrode, indicating secondary reactions of polybromine intermediates and residual Br₂, which can lead to energy loss and side reactions. In contrast, this peak was significantly reduced or disappeared for NZ-aNC@Br₂, showing a more efficient and simplified conversion process. Moreover, SEM images after cycling confirmed that the anode surface in the NZ-NC@Br₂ battery showed severe corrosion due to bromine shuttle, while the anode in the NZ-aNC@Br₂ battery remained intact, further validating the efficient confinement of bromine intermediates at the cathode in the confined structure (Fig. 37i). From a structural mechanism perspective, the unsaturated coordination sites introduced by the precursor activation in NZ-aNC construct abundant defect sites (Fig. 37j), which can modulate the local electronic state distribution, enhancing the interfacial interaction between active species and the carbon support. The abundant microporous structure, highly compatible with the size of bromine species, facilitates reversible anchoring and conversion reactions. The exposed metal residues also promote the graphitization of the carbon framework, improving conductivity and reducing polarization. This synergistic design demonstrates that the combination of metal-guided doping and functionalized strategies can effectively endow carbon hosts with multifunctional interfacial characteristics, providing a feasible material construction paradigm for high-performance, stable zinc–halogen energy storage systems.

Fig. 37 (a) Schematic of the synthesis of PNC-I₂. (b) Distribution of reduction energies for the dissociation of I₂ into I₃⁻ on graphene, pyridinic N, pyrrolic N and graphitic N. (c) Dissociation kinetics of I₃ on graphene, pyrrolic N, pyridinic N, and graphitic N. (d) UV-vis spectra of I₃⁻ supernatants after interaction with porous active carbon (PAC), PNC-900, and PNC-1000.²¹⁶ Reproduced with permission from ref. 216. Copyright 2022, Elsevier. (e) and (f) In situ UV-vis spectra and (g) and (h) in situ Raman spectra of NZ-NC@I₂ and NZ-aNC@I₂ during the charge–discharge process. (i) SEM images of NZ-NC and NZ-aNC anodes after 500 cycles. (j) Schematic of the energy storage mechanisms of NZ-NC and NZ-aNC cathodes in Zn–I₂ batteries.²¹⁷ Reproduced with permission from ref. 217. Copyright 2025, Wiley-VCH.

Compared to porous carbon materials, carbon felt electrodes offer higher mechanical strength and flexibility, allowing them to maintain structural integrity during charge and discharge cycles, preventing deformation and breakage, making them particularly suitable for electrochemical applications requiring high stability.²¹⁸ Additionally, the fibrous structure of the carbon felt provides excellent conductivity, enhancing the battery's power density. For example, Lu et al.⁴⁷ designed a multifunctional carbon felt electrode (NTCF) for Zn–Br₂ flow batteries (Fig. 38a). This electrode incorporates a nitrogen-deficient structure, significantly improving the catalytic activity for the Br₂/Br⁻ redox reaction and the regulation of zinc deposition while retaining the original conductive framework (Fig. 38b). The surface's rich defect sites increase the specific surface area and hydrophilicity, helping to reduce diffusion polarization and enhance the efficiency of interfacial reactions (Fig. 38c). DFT calculations (Fig. 38d) show that this material not only exhibits strong adsorption of Br₂ molecules but also effectively regulates the deposition behavior of zinc atoms. Among the nitrogen types, graphite-like nitrogen, with its C–N bond, has the strongest adsorption effect on bromine molecules, while pyridine and pyrrole nitrogen create low-energy adsorption sites via adjacent vacancies, where unpaired electrons can strongly interact with Br₂'s lone pair. Additionally, zinc atoms preferentially deposit in nitrogen-rich vacancy regions, helping to form a compact, uniform zinc layer and suppress dendrite growth. This structure enhances the catalytic efficiency of the Br₂/Br⁻ reaction, achieving stable confinement and conversion of active species, thereby granting the Zn–Br₂ battery excellent areal capacity and rate performance (Fig. 38e). In addition to defect-engineered carbon felt materials that co-regulate the Br₂/Br⁻ conversion path with a conductive framework, researchers have explored the introduction of redox mediators to precisely control the reaction path. Zhang et al.²¹⁹ introduced Prussian blue-modified nitrogen-doped carbon materials (PB@NC) as an auxiliary redox-targeted catalyst to address the slow bromine redox reaction kinetics and severe self-discharge in Zn–Br₂ systems. In the traditional cathode reaction pathway, bromide ions lose electrons during charging to form elemental bromine, a process with a high reaction free energy that limits the reaction rate (Fig. 38f). However, with the introduction of PB@NC, Prussian blue is first oxidized to Berlin green, with a significantly lower reaction energy barrier than the traditional pathway; then, Berlin green undergoes further redox reactions with bromine ions, generating elemental bromine and regenerating Prussian blue, allowing for the cyclical regulation of the redox mediator. DFT calculations confirm that with the involvement of BG (Berlin Green)/PB, the bromine oxidation path is divided into two consecutive reactions with lower energy barriers, effectively reducing the reaction energy demand. Moreover, analysis using the Nernst equation shows that this targeted catalytic pathway maintains good spontaneity during discharge and can continue to drive the reaction in the later stages of battery operation, enhancing the conversion rate and confinement capability of intermediate bromine species (Fig. 38g). The nitrogen-doped carbon framework in PB@NC provides stable adsorption sites for halogen species, while the redox mediator precisely regulates the reaction path and energy barriers, demonstrating excellent confinement and catalytic effects in suppressing active species shuttling and facilitating rapid bromine species conversion, significantly improving the electrochemical kinetics and operational stability of Zn–Br₂ batteries.

Fig. 38 (a) Preparation of the NTCF electrode and its working mechanism in ZBFBs. (b) Adsorption energy of Br₂ on graphite N, pyridinic N, and pyrrolic N sites of NTCF surfaces. (c) Optimized structure of the bromine molecule adsorbed on the pyrrolic N surface and the reaction mechanism of Br₂/Br⁻ coupling on NTCF. (d) Adsorption energy of zinc atoms on graphite N, pyridinic N, and pyrrolic N sites of NTCF surfaces. (e) Comparison of current density and areal capacity between ZBFB and previously reported systems.⁴⁷ Reproduced with permission from ref. 47. Copyright 2021, Wiley-VCH. (f) Conventional and newly proposed mechanisms of the bromine cathode. (g) Redox potential evolution of BG/PB and Br₅⁻/Br⁻ during the discharge process.²¹⁹ Reproduced with permission from ref. 219. Copyright 2024, Elsevier.

Recent studies have demonstrated the application of doping strategies to hierarchical carbon materials, achieving efficient halogen species confinement and multi-scale enhancement of electrochemical reactions through the synergistic optimization of both structure and chemistry. For example, Lee et al.⁴² reported a nitrogen-doped hierarchical porous carbon (NGF) as a host material, which was fabricated by oxidizing GF, followed by the in situ growth of ZIF-8 on its surface, and finally pyrolyzing the material under an argon atmosphere to form nitrogen-doped carbon (Fig. 39a). GF, commonly used in flow battery electrodes, features a large pore channel structure that facilitates rapid ion migration. The introduction of ZIF-8 precursors enabled the in situ formation of microporous carbon, thus creating a membraneless and flowless (MLFL) Zn–Br₂ battery (MLFL-ZBB) system with excellent bromine species conversion and storage capabilities (Fig. 39b). To further understand the adsorption mechanism of bromine species by nitrogen-doped carbon, the authors conducted systematic DFT calculations (Fig. 39c), comparing the adsorption behavior of different bromine species on monolayer graphene and nitrogen-doped graphene surfaces. The results showed that nitrogen doping, especially hydrogenated pyridine nitrogen (hPL) under acidic conditions, significantly enhanced the adsorption of Br⁻ and polybromide ions (Br₃⁻, Br₅⁻, and Br₇⁻), increasing the interfacial binding strength and helping to stabilize intermediate species. However, the enhancement for Br₂ adsorption was weaker, indicating higher selectivity for anionic intermediates. To further reveal the relationship between structure and performance, the authors proposed a reaction mechanism model (Fig. 39d). Despite NGF-1000 having a certain microporous structure capable of confining polybromide ions less than 2 nm, it still faced intermediate leakage risks due to insufficient strong adsorption sites. In contrast, NGF-700, at a lower carbonization temperature, formed a nitrogen-rich defect microporous structure, significantly enhancing interfacial adsorption and conversion abilities, effectively suppressing the diffusion and loss of polybromine species. This study suggests that by regulating carbonization temperature and nitrogen source configuration, porous carbon cathodes with high selectivity for bromine species adsorption and stable confinement ability can be developed, providing a structural design approach for efficient and stable Zn–Br₂ batteries. Compared to traditional non-metallic heteroatom doping strategies, which primarily rely on introducing lone pair electrons or constructing polar active sites to regulate iodine species adsorption behavior, halogen elements themselves possess high electronegativity and unique p-orbital electronic structures, making them better suited for electronic coupling and chemical affinity with iodine species. Therefore, directly introducing halogen elements for in situ doping is expected to provide superior control in confinement adsorption, polyhalide ion conversion, and interfacial electron transfer. Recently, Cao et al.²²⁰ constructed a bromine-doped carbon-nitrogen nanosheet (BrCN) and proposed it as a novel iodine cathode host material with excellent adsorption properties and rapid reaction kinetics. Charge density difference maps showed electron depletion near the triazine ring (cyan) and electron accumulation around iodine species (yellow), indicating significant charge exchange between the host and iodine species (Fig. 39e), facilitating the formation of a stable chemical adsorption structure. The introduction of Br atoms modulated the material's electronic structure, making BrCN exhibit stronger binding ability for iodine species and polyiodide ions (Fig. 39f). Compared to undoped materials, BrCN showed lower adsorption energy and stronger electronic coupling ability for polyiodide species, facilitating effective confinement and anchoring at the microscopic scale. Meanwhile, Br doping significantly optimized the material's energy band structure, enhancing its electronic migration efficiency and faster reaction kinetics in interfacial reactions with iodine species (Fig. 39g). Free energy analysis further indicated that BrCN has a greater driving force in the reduction path of iodine, promoting the efficient conversion of I₂ to I⁻ (Fig. 39h). Combined with its 2D porous structure, BrCN provides abundant active sites and a large specific surface area, not only enhancing the confinement effect of iodine species but also significantly improving the reaction kinetics and cycling stability of the iodine cathode (Fig. 39i), showing its broad application potential in constructing high-performance iodine-based energy storage systems.

Fig. 39 (a) Synthetic route of NGFs. (b) Schematic of the structure of MLFL-ZBB. (c) Adsorption energies of Br⁻, Br₂, Br₃⁻, Br₅⁻, and Br₇⁻ on graphene and hydrogenated pyridine-substituted graphene. (d) Reaction mechanisms of NGF-700 and NGF-1000 electrodes.⁴² Reproduced with permission from ref. 42. Copyright 2019, Wiley-VCH. (e) Top and side views of charge density difference maps for BrCN-I₃⁻, BrCN-I₂, and BrCN-I⁻. (f) Calculated adsorption energy of I₂, I₃⁻, and I⁻ on CN and BrCN. (g) PDOS plots of I⁻, I₂, and I₃⁻ species adsorbed on BrCN and CN. (h) Gibbs free energy profiles of BrCN and CN during the discharge process. (i) Schematic of the energy storage mechanism of the BrCN host.²²⁰ Reproduced with permission from ref. 220. Copyright 2024, American Chemical Society.

Compared to single-atom doping, co-doping can further enhance the confinement capability of carbon materials for iodine species due to the local charge redistribution on the surface of carbon materials, resulting in multiple synergistic effects during the electrochemical process.²²¹ However, research on the underlying mechanisms remains limited, and systematic understanding is still lacking. In this context, Feng et al.²²² designed and synthesized an N, S co-doped porous carbon material (denoted as NS-YP80F) as a host for iodine species (Fig. 40a). Kinetic analysis revealed that NS-YP80F@I₂ exhibited the lowest activation energy, indicating its optimal electrocatalytic kinetics during the I₂/I⁻ conversion process (Fig. 40b). Further Gibbs free energy calculations also confirmed the catalytic advantages of this material in the iodine species conversion pathway (Fig. 40c). Although nitrogen doping can enhance the electronegativity of the carbon matrix, thereby improving its binding ability with iodine species, its catalytic activity remains limited in the conversion of intermediate species such as I₃⁻/I₅⁻. In contrast, N, S co-doping effectively promotes charge transfer and lowers the reaction energy barrier, demonstrating superior catalytic kinetics (Fig. 40d). In situ Raman spectroscopy further revealed the intermediate evolution behavior during the electrochemical conversion process (Fig. 40e and f). At the discharge cut-off potential (0.8 V), both YP80F@I₂ and NS-YP80F@I₂ electrodes showed no obvious I₃⁻/I₅⁻ peaks. However, during charging, the peaks at 110 cm⁻¹ and 163 cm⁻¹ gradually increased and rapidly decreased when charged to 1.8 V, indicating that the intermediate polyiodine species were efficiently converted to I₂, with good reversible redox properties. In contrast, the YP80F@I₂ electrode retained the I₃⁻/I₅⁻ signals at higher potentials, reflecting incomplete intermediate conversion and lower reversibility and efficiency. These results systematically demonstrate the synergistic advantage of nitrogen-enhanced sulfur active sites in NS-YP80F, which suppress polyiodine anion shuttling and promote efficient iodine species conversion, providing new insights for the material design of high-performance zinc–halogen batteries. Although most studies easily tune the doping level by adjusting the calcination temperature, the doping levels are generally low, which limits the ability to bind polyiodide species and the adsorption performance for zinc ions. Therefore, the development of co-doped porous carbon cathode materials with high doping content remains a scientific challenge.

Fig. 40 (a) Mechanisms of the pristine porous carbon host, S-doped porous carbon host, and N-enhanced S-doped porous carbon host in Zn–I₂ batteries. (b) Arrhenius plots of Zn–I₂ batteries with different electrodes: YP80F@I₂, S-YP80F@I₂, N-YP80F@I₂, and NS-YP80F@I₂. (c) Gibbs free energy diagrams of the I₂ reduction reaction in batteries using NS-YP80F@I₂, N-YP80F@I₂, S-YP80F@I₂, and YP80F@I₂ cathodes, respectively. (d) Schematics of the working mechanisms of NS-YP80F@I₂, S-YP80F@I₂, and YP80F@I₂ electrodes, respectively. In situ Raman spectra of the (e) NS-YP80F@I₂ cathode and (f) YP80F@I₂ cathode at various charge–discharge voltages.²²² Reproduced with permission from ref. 222. Copyright 2024, Elsevier.

In summary, the introduction of heteroatoms such as nitrogen, sulfur, and phosphorus significantly alters the electronic structure and surface chemical environment of carbon materials, enhancing the adsorption capacity and redox reaction kinetics of iodine/bromine species. Among these, graphitic-type nitrogen, defect-rich structures, and dual-atom synergistic sites demonstrate excellent performance in improving reaction reversibility, suppressing intermediate accumulation, and mitigating the shuttle effect. In particular, the combination of dual-atom doping and functional interface catalysis (such as PB-assisted pathway regulation) strategies has provided new synergistic control approaches for enhancing the confinement and conversion efficiency of halogen species. However, several challenges remain to be addressed: (i) the configuration regulation and active mechanisms of heteroatom doping sites are still unclear, and there is a lack of systematic structure–activity relationship analysis and precise control strategies; (ii) while some doping methods improve catalytic performance, they may lead to a decline in electrical conductivity or structural stability, making it difficult to achieve a balanced and synergistic optimization of performance; (iii) existing research is predominantly focused on the iodine system, and the confinement mechanisms and electrochemical conversion pathways for other halogen species such as bromine and chlorine are still insufficiently studied and require further expansion and in-depth investigation. Future research should further strengthen the synergistic application of theoretical calculations and in situ characterization techniques, systematically revealing the effects of heteroatom doping on charge distribution, interface reaction pathways, and intermediate state behaviors. Additionally, multi-element synergistic doping strategies can be developed to regulate the electronic structure and interface chemistry of halogen-based cathodes, enabling an efficient confinement–catalysis–conduction process of halogen species, enhancing their broad adaptability in zinc–halogen batteries.

4.4. Atomic-level engineering of SACs/DACs

Atomic-level engineering of SACs and DACs represents a promising strategy to enhance the performance of Zn–I₂ batteries.³¹ By optimizing the dispersion and coordination of metal atoms, SACs/DACs can significantly improve the conversion of iodine species, stabilize halogen species, and enhance cycling stability, effectively addressing issues such as polyiodide shuttling and sluggish redox kinetics.²²³ For example, Ma et al.²²⁴ designed and constructed a multi-level porous carbon material (NiSAs-HPC) loaded with single-atom nickel through a template method (Fig. 41a). XANES and EXAFS characterization confirmed that the Ni atoms in the material exist in the +2 oxidation state and are predominantly single-dispersed on the carbon framework via Ni–N coordination (Fig. 41b and c). Further in situ Raman spectroscopy studies (Fig. 41d) on NiSAs-HPC in a Zn–I₂ battery revealed that it displayed typical I₃⁻ and I₅⁻ characteristic peaks, which gradually decreased until disappearing during discharge, indicating the reversible conversion of polyiodide anions (I₅⁻ → I₃⁻ → I⁻) within the NiSAs-HPC host. This showed that the abundant microporous and mesoporous structure of NiSAs-HPC effectively inhibited the shuttle effect of polyiodide anions, achieving efficient confinement (Fig. 41e). The confined Ni single atoms, acting as electrocatalytic active sites, significantly accelerated the redox reaction rate of iodine, demonstrating an excellent synergy between confinement and catalysis (Fig. 41f). Driven by these two mechanisms, even with a high mass loading of 11.6 mg cm⁻², the Zn–I₂ battery assembled with this material exhibited outstanding capacity stability over long cycles, with a cycle life exceeding 10 000 cycles. To further explore the synergistic mechanism of single-atom catalytic sites in inhibiting halogen shuttle and accelerating multi-step redox reactions, Qu et al.²²⁵ developed a highly dispersed single-atom Ni catalyst supported on a porous carbon fiber substrate (Ni–N₄CNF), serving as an active center with both confinement and dual-catalysis functions in a Zn–I₂ battery. By precisely tuning the coordination structure of Ni, this material effectively suppressed iodine species’ shuttle effect while significantly improving the reaction kinetics of I₂/I₅⁻/I₃⁻/I⁻ (Fig. 41g). Compared to traditional pure carbon materials (Fig. 41h), the Ni–N₄ sites exhibited stronger chemical anchoring and higher electrocatalytic activity, preventing iodine loss and Zn anode corrosion, thus enhancing the battery's cycle life and energy efficiency. Charge density difference analysis (Fig. 41i) showed that after adsorbing I₂ and I₃⁻, the Ni–N₄ sites exhibited significant charge accumulation, enhancing electrostatic interactions with iodine species, thereby improving their anchoring and activation ability. Additionally, the evident elongation of the I–I bond further suggested that this structure facilitates iodine molecule cleavage and recombination, optimizing its reaction pathway. Gibbs free energy calculations revealed that the reaction energy barriers for I₂ → I₅⁻ and I₅⁻ → I₃⁻ on the Ni–N₄ substrate were lower than those on pure carbon materials, reflecting its excellent catalytic effect (Fig. 41j). Thanks to the structural advantages of Ni–N₄CNF, the electrode demonstrated outstanding discharge performance, high energy efficiency, and excellent cycling stability at high rates. The above studies indicate that the synergistic effect of porous structures and SACs not only enhances the confinement and adsorption capacity of iodine species but also accelerates their conversion reactions through atomic-level catalytic activity, while constructing an efficient electron/ion conduction network, providing promising strategies for effective halogen-confining host materials.

Fig. 41 (a) Fabrication process of NiSAs-HPC. (b) Ni K-edge XANES spectra and (c) the corresponding FT-EXAFS spectra of NiSAs-HPC, NiO, and Ni foil. (d) In situ Raman spectra of the NiSAs-HPC/I₂ cathode during charge–discharge processes. (e) Charge–discharge process of Zn–I₂ batteries based on the NiSAs-HPC/I₂ cathode. (f) Schematic of the key functions of NiSAs.²²⁴ Reproduced with permission from ref. 224. Copyright 2023, American Chemical Society. Comparative schematic of reaction mechanisms between the (g) pure carbon fiber cathode and (h) Ni single-atom cathode. (i) Differential charge density of pure carbon and Ni–N₄C. (j) Gibbs free energy profiles for the conversion of I₂ to I₃⁻ and I₅⁻ on pure carbon and Ni–N₄C substrates.²²⁵ Reproduced with permission from ref. 225. Copyright 2024, Wiley-VCH.

Similar to Ni, Co single atoms can provide highly active sites for iodine species conversion, but they offer distinct advantages due to their unique electronic properties and coordination environments.¹⁹¹ In this regard, Co SACs supported on carbon are also being explored for Zn–I₂ batteries, further expanding the possibilities for improving battery performance. For example, Zhu et al.²²⁶ reported a nanoreactor structure with a capture–adsorption–catalysis multifunctional synergy, where electron-rich cobalt nanoparticles (Co NPs) are uniformly embedded in a porous activated carbon matrix (Co@AC) (Fig. 42a). Experimental results showed that Co@AC could rapidly decolorize and significantly reduce the concentration of I₃⁻ in both direct contact with I₂ solution and reaction with Zn(I₃⁻)₂ solution. UV-vis absorption spectra clearly demonstrated a significant decrease in the absorption peak, indicating its excellent polyiodide anion confinement ability (Fig. 42b). XPS analysis (Fig. 42c) revealed that the binding energy of Co²⁺ in the Co@AC-I₃⁻ sample shifted to a lower energy level, suggesting strong chemical interactions between the Co sites and polyiodide anions. This interaction facilitated interfacial charge transfer, enhancing the electrocatalytic ability for iodine species conversion. Moreover, to investigate its impact on Zn²⁺ transport, CV tests were performed, comparing the behavior of Co@AC/I₂ and AC/I₂ electrodes at various scan rates. Results showed that the Co@AC/I₂ electrode exhibited higher slopes and lower polarization throughout all redox processes, reflecting a faster Zn²⁺ diffusion rate and superior iodine redox kinetics (Fig. 42d and e). Kinetic barrier calculations further confirmed that, compared to the AC/I₂ electrode, the Co@AC/I₂ electrode demonstrated significantly reduced energy barriers in the I₂ reduction and I⁻ oxidation processes (Fig. 42f), which is attributed to the synergistic enhancement of spatial confinement and chemical adsorption of iodine species by the cobalt nanoparticle sites in the material. To further explore the reaction pathways and shuttle behavior of iodine species, the researchers conducted multi-angle in situ analysis (Fig. 42g), comparing the evolution at the cathode, electrolyte, and Zn anode interfaces for Co@AC/I₂ and AC/I₂ electrodes. In situ UV-vis results showed that during charging, the I₃⁻ intensity in the traditional AC/I₂ electrode system rapidly increased, indicating severe iodine dissolution (Fig. 42h). In contrast, in the Co@AC/I₂ system, I₃⁻ remained at low concentrations, indicating effective inhibition of polyiodide anion dissolution and shuttle migration (Fig. 42i). The electrode demonstrated outstanding rate performance, achieving a capacity of 221.1 mAh g⁻¹ at 0.5 C, 89% energy efficiency at 50 C, and 125.9 mAh g⁻¹ discharge capacity at 2.5 C in a soft-pack battery, fully demonstrating the critical role of the confinement-catalysis structure in suppressing iodine shuttle, enhancing reaction kinetics, and improving energy utilization. Compared to single-atom catalytic mechanisms, doping elements can modulate charge density, electronic orbital overlap, and reaction pathways at the atomic scale, enabling multi-dimensional catalytic enhancement.²²⁷ Based on this, Sun et al.²²⁸ systematically evaluated the performance of their single-atom cobalt-anchored nitrogen-doped porous carbon material (Co-SAs@NPC) for iodine species confinement. UV-vis absorption spectra results showed (Fig. 42j) that Co-SAs@NPC exhibited significantly stronger adsorption capability for the typical triiodide anion (I₃⁻), displaying far superior decolorization effects and absorption features compared to the undoped cobalt control sample (NPC), verifying its excellent confinement performance. DFT calculations indicated that iodine species such as I₂, I⁻, I₃⁻, and I₅⁻ had negative adsorption energies on the Co active sites, suggesting thermodynamic spontaneity and chemical anchoring (Fig. 42k and l). At a current density of 5 A g⁻¹, the material maintained good long-cycle performance (Fig. 42m), surpassing that of most previously reported Zn–I₂ battery systems (Fig. 42n). This fully validates the effectiveness of the single-atom cobalt and nitrogen doping synergistic design strategy for enhancing iodine species confinement adsorption and efficient conversion in zinc–halogen batteries.

Fig. 42 (a) Mechanism of an aqueous Zn–I₂ battery and its associated challenges. (b) Optical images and the corresponding UV-vis spectra of AC and Co@AC after soaking in I₃⁻ solution for 24 h. (c) Co 2p XPS spectra of Co@AC-I₃⁻ and Co@AC. CV profiles at various scan rates of button cells assembled with (d) Co@AC/I₂ and (e) AC/I₂ electrodes. (f) Activation energies during the reaction process for AC/I₂ and Co@AC/I₂ electrodes. (g) Characterization techniques for probing the internal reaction mechanisms of batteries. In situ UV-vis spectra of the electrolyte during charging in batteries using (h) AC/I₂ and (i) Co@AC/I₂ cathodes, respectively.²²⁶ Reproduced with permission from ref. 226. Copyright 2024, Wiley-VCH. (j) Comparison of UV-vis spectra of triiodide solutions before and after adsorption on NPC and Co-SAs@NPC. (k) Comparison of triiodide adsorption capacities between NPC and Co-SAs@NPC. (l) Adsorption energy of I₂, I⁻, I₃⁻, and I₅⁻ on Co sites. (m) Cycling performance of the Co-SAs@NPC/I₂ cathode. (n) Comparison of performance of the Co-SAs@NPC/I₂ cathode with previously reported iodine host cathodes.²²⁸ Reproduced with permission from ref. 228. Copyright 2024, Elsevier.

Compared to iodine species, bromine species exhibit differences in oxidation–reduction pathways and intermediate behaviors.⁴¹ To enhance the electronic regulation and interface stability of host materials, researchers have explored combined Co single-atom and N doping strategies to improve the performance of Zn–Br₂ batteries. For example, Li et al.²²⁹ reported a cobalt–nitrogen co-doped carbon material-modified graphite felt electrode (Co–N/C@GF) as a host material for halogen species confinement (Fig. 43a). This electrode demonstrates excellent bromine species adsorption capacity, allowing the battery to maintain an open-circuit voltage for up to 84 hours, significantly delaying the self-discharge process (Fig. 43b). This is attributed to the cobalt sites in the Co–N/C@GF, which provide abundant catalytic active centers for the bromine redox reaction, thereby effectively reducing electrochemical polarization and enhancing reaction kinetics. To further investigate the interaction between single-atom catalytic sites and halogen species, Yang et al.¹⁹¹ combined DFT theoretical calculations with synchrotron radiation characterization techniques to systematically study the role of various metal-N₄ SACs in the confinement adsorption and catalytic process of I₂ and its intermediates (Fig. 43c). Theoretical calculations revealed that the I₂ adsorption at the Cu sites in the single-atom copper (SACu) model was the most stable, superior to N and C sites, and the redox process at the Cu site exhibited the lowest energy barrier (Fig. 43d and e), demonstrating the thermodynamic advantage and catalytic potential of this site. Synchrotron XAFS techniques were then employed to characterize the structure and electronic state of single-atom copper embedded nitrogen-doped Ketjen black (SACu@NKB), revealing that Cu atoms were stably dispersed in the carbon framework in the Cu–N coordination form, with no detectable Cu–Cu pairing signals, confirming the single-atom dispersion feature (Fig. 43f–h). Further assessment of its I₃⁻ adsorption performance showed that, when equivalent amounts of nitrogen-doped Ketjen black (NKB) and SACu@NKB were added to a triiodide anion solution, the former caused almost no color change, while the latter rapidly decolorized the solution (Fig. 43i). UV-vis absorption spectra further confirmed that the characteristic absorption peaks of SACu@NKB at 288 and 350 nm significantly decreased, showing stronger adsorption than single atom Co-embedded nitrogen-doped Ketjen black (SACo@NKB) and pure NKB, validating its strong confinement adsorption capacity for I₃⁻ (Fig. 43j and k). Raman results showed that after the first cycle, the NKB electrode exhibited significant I₃⁻ penetration through the separator; in contrast, SACo@NKB alleviated this issue to some extent, while no I₃⁻ signals were detected in the separator of SACu@NKB within a 55 μm range, indicating its most significant effect in suppressing the iodine shuttle. CV tests revealed that the SACu@NKB electrode exhibited the highest reduction peak current, significantly outperforming SACo@NKB and unmodified NKB electrodes (Fig. 43l), indicating stronger catalytic activity in iodine species electrocatalytic conversion. Additionally, SACu@NKB displayed the lowest Tafel slope and activation energy (Fig. 43m and n), further demonstrating its excellent reaction reversibility and rapid electrochemical dynamics. Electrochemical performance tests further confirmed that SACu@NKB outperformed comparison samples in specific capacity, reaction kinetics, and cycling stability, fully reflecting its synergistic advantages in confinement adsorption and electrocatalysis. This work establishes a representative single-atom confinement-catalysis system, which presents a structural paradigm applicable to zinc–halogen batteries.

Fig. 43 (a) Schematic of the synthesis procedure for the Co-N/C@GF composite. (b) Open-circuit voltage profiles over time for ZBFBs with GF, HT@GF, and Co–N/C@GF electrodes.²²⁹ Reproduced with permission from ref. 229. Copyright 2024, Elsevier. (c) Interaction models between I₂ molecules and the active sites of C, N, and metal atoms in SACs@NG. (d) Adsorption energy of I₂ on C, N and SACu active sites. (e) Gibbs free energy profiles for I₂ reduction C, N and SACu active sites. (f) Normalized Cu K-edge XANES spectra and (g) FT-EXAFS spectra of SACu@NKB, Cu₂O, and Cu foil. (h) Fitted FT-EXAFS curves of SACu@NKB at the Cu K-edge. (i) Photographic comparison of triiodide adsorption by NKB, SACo@NKB, and SACu@NKB, and (j) thee corresponding UV-vis spectra of the solutions. (k) Specific adsorption capacity of NKB, SACo@NKB, and SACu@NKB for triiodide. (l) CV curves and (m) the corresponding Tafel plots of Zn–I₂ batteries assembled with NKB, SACo@NKB, and SACu@NKB cathodes, respectively. (n) Activation energy plots of NKB, SACo@NKB, and SACu@NKB cathodes.¹⁹¹ Reproduced with permission from ref. 191. Copyright 2023, Royal Society of Chemistry.

Compared to Cu SACs, Fe-based SACs offer a distinct advantage due to their highly tunable oxidation states (Fe²⁺/Fe³⁺) and abundant uncoordinated d orbitals, which allow for greater versatility in electronic structure regulation and facilitate improved catalytic performance.²³⁰ Based on this, studies have attempted to incorporate Fe-based SACs into carbon-based frameworks to construct efficient confinement-catalysis systems. For instance, Chen et al.¹⁸³ designed a single-atom iron site dispersed in a mesoporous carbon framework (CMK) as a confinement–catalysis structure (FeSAC-CMK). High-angle annular dark field STEM (HADDF-STEM) images revealed the presence of numerous uniformly distributed bright spots in the material, confirming the atomic dispersion of Fe (Fig. 44a). XANES spectra indicated that the K-edge absorption energy of Fe in the material was higher than that of Fe foil, suggesting that its valence state was between 0 and +1 (Fig. 44b); EXAFS Fourier transform spectra showed a clear Fe–N coordination peak at 1.5 Å, with no Fe–Fe signal observed around 2.2 Å, further verifying its single-atom structure (Fig. 44c). In situ Raman spectroscopy analysis (Fig. 44d) showed that, during charge and discharge, the characteristic vibration peak of Br₃⁻ in FeSAC-CMK gradually weakened during discharge and did not increase during charge, indicating the effective suppression of the regeneration of soluble Br₃⁻ and the reaction path was more inclined to Br⁻/Br⁰ conversion. In contrast, the comparative electrode without single-atom Fe showed a decrease in the Br₃⁻ peak during discharge, with a subsequent increase during charge, reflecting the continuous generation and shuttle behavior of Br₃⁻. XPS spectra (Fig. 44e) further confirmed that the Br 3d peak of FeSAC-CMK shifted to lower binding energies during discharge to 0.8 V and to higher binding energies during charge to 1.85 V, suggesting that the reduced product was Br⁻ and that no Br₃⁻ was generated during oxidation, which was more likely Br⁰ species anchored through the FeN₅ sites. DFT calculations (Fig. 44f) showed that FeSAC-CMK has a lower reaction energy barrier in the Br⁻ oxidation process, significantly promoting the formation of Br⁰. Additionally, Br⁰ is more stable in this material, and its conversion to Br₃⁻ requires overcoming an energy barrier, effectively suppressing the rapid formation of unstable intermediates. In contrast, on unmodified CMK, Br⁰ easily converts to Br₃⁻, while FeSAC-CMK is more favorable for the confinement and enrichment of Br⁰, optimizing the reaction path and enhancing the reaction kinetics. Wannier function analysis revealed strong π/σ overlap between the Fe(d)–Br(p) orbitals, lowering their energy levels and increasing the energy barrier for Br₃⁻ conversion (Fig. 44g), significantly enhancing the chemical anchoring ability of Br⁰ at the Fe sites, thereby improving the utilization of active species. The Zn||Br₂ battery constructed with FeSAC-CMK as the cathode achieved a specific capacity of 344 mAh g⁻¹ at 0.2 A g⁻¹ (Fig. 44h), fully demonstrating the excellent bromine confinement–catalysis performance and energy storage potential of this material. To further expand the structural types and regulation mechanisms of single-atom catalysis systems, recent research has focused on the potential regulation capabilities of non-traditional transition metals (such as Zn) in single-atom catalysis, aiming to construct carbon-based host materials with novel electronic structure characteristics. For example, Pei et al.²³¹ designed a carbon host material derived from edible fungal residues and constructed a composite structure with zinc single-atom catalytic sites (SAZn@C_FS), offering dual functionalities of confinement and catalysis for iodine species. UV absorption analysis and adsorption capacity tests showed that SAZn@C_FS exhibited excellent adsorption capacity for I₃⁻, significantly outperforming traditional activated carbon materials (e.g., YP80F and CNT) and the C_FS host without zinc single atoms, indicating that the introduction of zinc single atoms significantly enhanced the binding capacity for polyiodide anions (Fig. 44i). DFT calculations showed that SAZn@C_FS exhibited higher adsorption energies for I⁻, I₃⁻, I₅⁻, and I₂, accompanied by significant charge transfer behavior, suggesting that Zn–N₄ sites can serve as electronic regulation and catalysis centers, effectively promoting the conversion of iodine species and suppressing intermediate shuttling (Fig. 44j). In situ Raman spectroscopy (Fig. 44k) further revealed the conversion path of polyiodide intermediates during charge and discharge in different iodine hosts, where in the SAZn@C_FS system, the characteristic peaks of I₃⁻ and I₅⁻ were almost undetectable at the end of charging, indicating that under the synergistic effect of high pyridine nitrogen content and Zn–N₄ sites, I⁻ was rapidly and thoroughly converted to I₂. Differential charge density analysis (Fig. 44l and m) showed that the Zn–N₄–C site produced more significant electronic rearrangement than the N–C site when adsorbing I₃⁻, demonstrating stronger electronic coupling and chemical anchoring ability, which helped reduce the ΔG of the I⁻/I₂ full reaction path and the energy barrier for key steps (I⁻/I₃⁻ conversion), thereby effectively enhancing confinement catalysis and reaction kinetics. Gibbs free energy calculations showed that the reaction energy barriers for I⁻ → I₃⁻ and I₅⁻ → I₂ were significantly lower in SAZn@C_FS than that in C_FS, demonstrating better thermodynamic and kinetic reaction properties (Fig. 44n). Especially in the rate-limiting step of I⁻ → I₃⁻ (Fig. 44o), the reaction energy barrier of SAZn@C_FS was only 0.83 eV, much lower than the 1.03 eV for C_FS, further verifying the key role of zinc single atoms in promoting the I⁻/I₃⁻/I₅⁻/I₂ full-path conversion kinetics. Electrochemical performance tests showed that the iodine diffusion coefficient in SAZn@C_FS electrodes was as high as 10⁻⁹ cm² s⁻¹, reflecting excellent ion diffusion ability (Fig. 44p). In addition, the Zn–I₂ battery based on this material showed good cycling life, retaining 80% of its capacity after 3500 cycles at 3 A g⁻¹ (Fig. 44q). This study systematically revealed the synergistic mechanism between Zn single-atom sites and the hetero-porous carbon framework in the confinement adsorption and electrocatalytic process, effectively suppressing the dissolution and shuttle of polyiodide anions while significantly improving reaction reversibility and electrochemical stability. These results validate the universality of SACs in regulating mechanisms within multi-halogen systems, demonstrating the potential of efficient confinement–catalysis–conduction host material design in the development of high-performance rechargeable batteries.

Fig. 44 (a) HAADF-STEM image of FeSAC-CMK. (b) Normalized Fe K-edge XANES spectra of Fe foil and FeSAC-CMK. (c) Fourier transform extended X-ray absorption fine structure (FT-EXAFS) spectra of Fe foil and FeSAC-CMK. (d) CV curves of the Zn–Br₂ battery with the FeSAC-CMK electrocatalyst, along with in situ Raman spectra of the Br₂ cathode during charge and discharge. (e) Br 3d XPS spectra of the Br₂ cathode with FeSAC-CMK. (f) Energy profiles of Br oxidation and reduction reactions on CMK and FeSAC-CMK. (g) Wannier functions of Fe(d_z²)–Br(p_z) and Fe(d_yz)–Br(p_x) orbitals. (h) Rate performance of Zn–Br₂ batteries using CMK and FeSAC-CMK electrocatalysts, respectively.¹⁸³ Reproduced with permission from ref. 183. Copyright 2024, Wiley-VCH. (i) UV-vis spectra of triiodide solutions for CNT, YP80F, C_FS, and SAZn@C_FS. (j) Comparison of adsorption energy of I₂, I⁻, I₃⁻, and I₅⁻ on SAZn@C_FS and C_FS. (k) GCD profiles and the corresponding in situ Raman spectra of the Zn–I₂ battery based on the SAZn@C_FS electrode. Top and side view models of I₃⁻ interacting with (l) C_FS and (m) SAZn@C_FS. (n) Gibbs free energy profiles for the oxidation reaction of I⁻ on C_FS and SAZn@C_FS. (o) Energy barrier comparison for the I⁻ → I₃⁻ conversion on C_FS and SAZn@C_FS. (p) Diffusion coefficient of the SAZn@C_FS cathode determined by the GITT curves. (q) Cycling performance of Zn–I₂ pouch cells with SAZn@C_FS cathodes and the corresponding GCD curves after various cycle numbers.²³¹ Reproduced with permission from ref. 231. Copyright 2025, Wiley-VCH.

Compared with SAC modification, the synergistic effects of DACs enable porous carbonaceous host materials to exhibit significantly enhanced properties, providing multiple benefits to the electrochemical performance of zinc–halogen batteries. The dual-metal sites within DACs optimize the electronic configuration of the two metal centers through short- and medium-range electronic interactions and provide additional adsorption sites that cooperatively modulate the adsorption of intermediates.²³² This synergistic effect and the multifunctionality of two adjacent metal atoms enable the catalysts to exhibit excellent catalytic performance in complex reaction systems involving multiple intermediates.²³³ Therefore, the rational introduction of DACs into halogen-confining host materials is critical for achieving high-performance zinc–halogen batteries. Dong et al.²³⁴ designed a honeycomb-like zinc dual-atom catalyst (Zn DAC) confined within nitrogen-doped carbon nanosheets, serving as an efficient iodine host material (Fig. 45a). Electrochemical testing revealed that the Tafel slope of Zn DACs was markedly lower in both oxidation and reduction directions compared to single-atom zinc-doped materials (ZnNC) and undoped carbon substrates (NC), confirming the synergistic catalytic advantage of Zn DACs in enhancing the oxidation–reduction reaction kinetics of I₂/I⁻ (Fig. 45b). DFT further supported that Zn DACs formed stable chemical adsorption structures with iodine species such as I₂ and I₃⁻via Lewis acid–base interactions, facilitating effective charge transfer, thereby reducing the reaction energy barrier and accelerating the conversion kinetics of polyiodide anions (Fig. 45c). Compared to conventional single-atom catalytic sites or undoped materials, the DAC structure exhibited enhanced electronic coupling and superior catalytic reaction pathways (Fig. 45d), effectively suppressing iodine shuttle effects and improving the battery's cycling stability and reversible reaction performance. This study, by constructing highly active Zn DACs, achieved efficient confinement and catalytic conversion of iodine species, paving the way for novel design strategies in the development of high-performance Zn–I₂ battery electrocatalysts. Building upon the excellent confinement and catalytic performance of Zn DACs, researchers further investigated the role of multi-metal synergistic effects in regulating halogen conversion behavior. Hei et al.²³⁵ proposed a dual-atom catalytic structure with manganese-zinc bimetallic atoms anchored on a nitrogen-doped carbon substrate (MnZn–NC), used to load iodine species and enhance their oxidation–reduction reaction performance. In the experiment, temperature-dependent magnetization tests via zero-field cooling (ZFC) were performed on Mn–NC and MnZn–NC (Fig. 45e). The results showed that the Mn sites in MnZn–NC exhibited a higher effective magnetic moment, with an unpaired d electron count of 2.84, higher than the 2.64 of Mn–NC. Combined with DFT calculations, MnZn–NC exhibited a higher spin density, indicating a higher proportion of Mn³⁺ involved in reactions (Fig. 45f). In the tetragonal planar symmetrical coordination structure, Mn existed in Mn²⁺ (S = 3/2) or Mn³⁺ (S = 4/2) states. The increase in magnetic moment suggested an elevated proportion of Mn³⁺, which benefited the optimization and regulation of its electronic structure. Further PDOS analysis (Fig. 45g) revealed that MnZn–NC exhibited a stronger metallic character than Mn–NC. Upon cooperative coordination with Zn, it formed spontaneously spin-polarized electronic states at the Fermi level, significantly enhancing the catalyst surface's electron transfer ability toward reactive intermediates. Meanwhile, compared to Zn–NC, the dxy orbitals of Zn in MnZn–NC were closer to the Fermi level, enhancing the electrocatalytic activity of the Zn sites. Gibbs free energy calculations showed that MnZn–NC exhibited lower ΔG values in each step of the I₂ reduction reaction compared to Mn–NC and Zn–NC (Fig. 45h), indicating superior thermodynamic driving forces for the reaction. The spin polarization and spin exchange between Mn and Zn bimetals regulated the catalyst's electronic structure, improving the adsorption thermodynamics and redox kinetics of iodine species. This effectively suppressed the I₃⁻ shuttle and promoted reversible iodine conversion, significantly enhancing the electrochemical performance of Zn–I₂ batteries (Fig. 45i and j). These results verified that the synergistic effects of Zn DACs overcame the limitations of SACs by improving the formation and desorption of intermediates and that adjacent active sites facilitated rapid oxidation–reduction kinetics of polyiodides, offering new insights for the rational design of DACs and the construction of high-efficiency energy storage devices. On this basis, Chun et al.²³⁶ recently proposed a self-driving computational strategy for further screening high-performance catalysts in the zinc–halogen system. This strategy combined first-principles calculations with equivariant graph neural networks (GNN), conducting high-throughput screening among over 30 000 binary metal sites composed of different 3d transition metal elements and various coordination environments, systematically exploring their confinement catalytic potential in halogen-related reaction systems. Through an active learning mechanism balancing the exploration of new structures with the use of known high-activity structures, the GNN model was able to accurately identify composition–structure–performance relationships in different localized chemical environments and predict catalytic activity and reaction selectivity in key reactions such as the oxygen reduction reaction (ORR) and oxygen evolution reaction (OER). The computational results showed that metal pairs such as Co–Fe, Co–Co, and Co–Zn demonstrated excellent catalytic performance, with the predicted results aligning closely with existing experimental data, further confirming the effectiveness of this strategy in identifying optimal active sites and elucidating the role of confined structures (Fig. 45k–n). This study, integrating theoretical calculations with data-driven design, establishes a foundation for the development of host materials with high adsorption capacity and catalytic activity for high-energy-density, long-cycle-life zinc–halogen batteries.

Fig. 45 (a) Synthesis process and working mechanism of I₂@Zn₂NC. (b) Tafel plots of NC, ZnNC, and Zn₂NC. (c) Comparison of adsorption energy of I⁻, I₂, and I₃⁻ on NC, ZnNC, and Zn₂NC catalysts. (d) Gibbs free energy profiles of the I₂ reduction reaction on NC, ZnN₄, and Zn₂NC.²³⁴ Reproduced with permission from ref. 234. Copyright 2025, Royal Society of Chemistry. (e) Spin density and magnetic moment, number of unpaired 3d electrons, and M–T curves of Mn–NC and MnZn–NC. (f) PDOS of Mn 3d orbitals of MnZn–NC and Mn–NC. (g) PDOS of Zn 3d orbitals of MnZn–NC and Mn–NC. (h) Free energy profiles of the I₂ reduction reaction on Zn–NC, MnZn–NC, and Mn–NC. (i) Current challenges in Zn–I₂ batteries. (j) Reaction mechanism of dual-atom MnZn–NC sites in Zn–I₂ batteries. Overpotentials of (k) ORR and (l) OER and the Gibbs free energy of adsorption for key intermediates (O*, OH*, and OOH*).²³⁵ Reproduced with permission from ref. 235. Copyright 2024, Wiley-VCH. (m) Metal sites with surface overpotentials below the standard threshold. (n) Stability evaluation of SACs.²³⁶ Reproduced with permission from ref. 236. Copyright 2024, Springer Nature.

In summary, this section systematically highlights that the introduction of highly dispersed single/double metal atom sites in carbon-based hosts significantly enhances the confinement adsorption ability and electrocatalytic activity for iodine/bromine species. This effectively suppresses the shuttle effect of polyhalide ions and accelerates their reversible redox reactions. These atomically precise catalytic structures, leveraging multiple advantages such as electronic structure regulation, interfacial charge reconstruction, and reaction barrier reduction, have driven continuous breakthroughs in the specific capacity, energy efficiency, rate performance, and cycling stability of zinc–halogen batteries. Notably, the M–N₄ coordination centers in SACs demonstrate excellent versatility and design flexibility in adsorption energy, electronic coupling, and reaction path regulation. However, DACs face challenges in synthesis and exhibit complex catalytic mechanisms during the reaction process, with research in the zinc–halogen battery field still being in its infancy. Future research could further expand in the following directions: (i) developing precisely controlled atomic-level loading strategies to regulate the configuration and optimize the spatial distribution of metal atom sites; (ii) extending the confinement-catalysis design to multi-scale synergistic structures, such as atom sites coupled with mesoporous structures, dual-metal sites combined with 2D confinement channels, and other composite strategies, to achieve coordinated reaction pathway control; (iii) deepening the use of in situ spectroscopy and theoretical methods to real-time analyze the valence state evolution and intermediate dynamics of SACs and DACs in actual electrochemical processes, providing a systematic understanding of the structure–performance–mechanism relationships to build high-energy density and long-cycle zinc–halogen batteries. In conclusion, SACs and DACs confinement catalysis strategies provide strong material support for the next generation of high-performance aqueous energy storage systems.

4.5. Heterostructure engineering

Heterostructures, by integrating two or more materials with complementary properties,^237,238 can significantly enhance the performance of halogen-confining cathode hosts in zinc–halogen batteries. The unique interfacial interactions within these heterostructures provide abundant active sites for the adsorption and conversion of halogen species,²³⁹ thereby accelerating the kinetics of redox reactions. Moreover, the synergistic effects between different components effectively suppress the shuttle effect of halogen species, reduce self-discharge, and improve cycling stability.²⁴⁰ Consequently, rationally designed heterostructures can mitigate the dissolution and shuttle effects of halogen species, thereby enhancing the electrochemical performance of the battery. Recently, Zhang et al.²⁴¹ developed a graphene/polyvinylpyrrolidone (G/PVP) heterostructure as the iodine host and utilized ZnI₂ as the active material for anode-free Zn–I₂ batteries (Fig. 46a). The introduction of PVP effectively suppressed the shuttle effect of iodine species, and the high conductivity and structural stability of graphene enhanced electron transport and cycling stability, demonstrating typical confinement-catalysis synergy. As a result, the ZnI₂ cathode provided sufficient Zn²⁺ ions as the zinc source for the anode-free system and delivered high specific capacity output and cycling durability. To further investigate the cycling stability of high-valent iodine species, Kang et al.²⁴² proposed a heterostructure host material based on an electron withdrawal strategy to stabilize high oxidation-state I⁺ species in zinc-free aqueous electrolytes (ZnSO₄). Copper phthalocyanine (CuPc), an organic-metallic molecule, served as the functional unit. The pyrrole N atoms in CuPc possess lone pairs of electrons on the sp² hybridized orbitals, and upon coordination with Cu²⁺ to form the Pc complex, deprotonation occurred, making all N atoms in CuPc capable of charge transfer interactions with I⁺ ions (Fig. 46b). HRTEM images of CuPc@rGO and CuPc@CNT composites showed no aggregation of CuPc in the stacked rGO layers or the interconnected CNT network (Fig. 46c). When CuPc was constructed with graphene or CNTs to form an in-plane π–π stacking structure, this heterostructure significantly enhanced the electron migration from the carbon substrate to CuPc, improving the nucleophilicity of the pyridine N atoms and stabilizing high-valent I⁺ species (Fig. 46d). The differential charge density distribution map (Fig. 46e) revealed that electron depletion was concentrated in the graphene region (blue isosurface), while electron accumulation was concentrated around CuPc (yellow isosurface), thereby verifying the electron withdrawal behavior. The integral charge density along the z-axis (Fig. 46f) further quantified this charge transfer process, where the negative peak corresponded to the graphene region and the positive peak to the CuPc region, visually reflecting the electron transfer pathway from the carbon backbone to CuPc. The CuPc@CNT-ZnI₂ cathode system constructed with this heterostructure exhibited significant pseudocapacitive behavior, with the capacitive contribution continuing to rise as the scan rate increased (Fig. 46g), indicating that the charge storage process was predominantly governed by fast surface reactions. In situ Raman results (Fig. 46h) showed that during the first charging phase (0.6–1.6 V), the peaks corresponding to I₃⁻ and I₅⁻ gradually intensified, suggesting the gradual formation of polyiodide anions. During the second phase (1.6–1.8 V), these peaks rapidly decreased, indicating that most polyiodide species were further oxidized to I⁺. During the subsequent discharge process, the signals of I₃⁻ and I₅⁻ first increased and then decreased (1.8–0.6 V), revealing the reversible redox process between I₃⁻/I⁻ and I₅⁻/I⁻. Additionally, during charging at voltages higher than 1.4 V, a new peak at around 330 cm⁻¹ emerged and rapidly disappeared when the voltage dropped below 1.4 V, suggesting that this vibrational mode was closely related to the generation and confinement stability of I⁺. Accordingly, the CuPc@rGO-ZnI₂ cathode, due to the enhanced π–π conjugation between CuPc and rGO, exhibited higher initial specific capacity and good capacity retention (Fig. 46i). This study systematically demonstrated the confinement effect and electrocatalytic performance tuning capability of heterostructured carbon-based hosts in stabilizing high-valent iodine species from the perspectives of molecular orbital regulation and interfacial electron behavior. Similarly, in the Zn–Br₂ system, the confinement design based on heterogeneous interface engineering also showed significant effectiveness. Li et al.²⁴³ constructed a three-dimensional hierarchical structure of Ni/NiO heterojunction composite electrodes (Ni/NiO@GF), which effectively enhanced the Br₂/Br⁻ redox kinetics by in situ generating a core–shell structure on the surface of GF. DFT calculations showed that the adsorption energy of Br atoms on the bare graphite surface was relatively high (2.20 eV), indicating weaker adsorption; whereas on the Ni (−1.41 eV) and NiO (−0.10 eV) surfaces, particularly at the oxygen vacancies within NiO (−2.31 eV), the adsorption ability was much stronger (Fig. 46j). These oxygen vacancies were largely generated during the in situ oxidation of NiO, enhancing the interaction with Br atoms (Fig. 46k). Further analysis revealed that Br atoms preferentially occupied oxygen vacancies and bonded with Ni, thereby achieving adsorption–reduction cooperative catalysis on the NiO surface, effectively promoting the conversion of Br₂/Br⁻ and suppressing the bromine shuttle effect. The reaction mechanism was that Br atoms were first adsorbed on the NiO shell and then reduced to Br⁻ under its catalysis (Fig. 46l). The battery based on Ni/NiO@GF still maintained about 82% voltage efficiency and high energy efficiency after 300 cycles at a current density of 80 mA cm⁻² (Fig. 46m), with the maximum energy density reaching 125 Wh L⁻¹ (Fig. 46n). Meanwhile, the strong adsorption of Ni/NiO@GF to Br₂ significantly reduced the self-discharge rate of the battery (Fig. 46o), further validating its role in the confinement and stabilization of bromine species. These studies offer effective strategies for the development of advanced host materials through the precise design of heterostructures.

Fig. 46 (a) Schematic diagram of the ZnI₂ zinc-rich cathode with the G/PVP host and its advantages.²⁴¹ Reproduced with permission from ref. 241. Copyright 2022, Springer Nature. (b) Schematic of the lone pair electrons on nitrogen atoms in Pc and CuPc. (c) HRTEM images of CuPc@rGO. (d) Structure of CuPc and schematic of charge transfer in the iodine cathode. (e) Top and side view differential charge density maps of CuPc@rGO. (f) Local integrated differential charge density profile of CuPc@rGO along the z-axis. (g) Capacitive and diffusion-controlled capacity contributions of CuPc@rGO-ZnI₂ at various scan rates. (h) Charge–discharge profiles of CuPc@rGO-ZnI₂, alongside its Raman spectra at corresponding potentials. (i) Cycling performance of CuPc@rGO-ZnI₂ at 3.0 A g⁻¹.²⁴² Reproduced with permission from ref. 242. Copyright 2024, Elsevier. (j) Adsorption energy of Br₂ on different sites of Ni/NiO@GF, including graphite, Ni top, NiO, and NiO vacancies. (k) Optimized configuration of the Br atom adsorbed at the oxygen vacancy site on NiO. (l) Redox reaction mechanism on the Ni/NiO heterostructure. (m) Cycling performance of ZBFBs with the Ni/NiO@GF cathode. (n) Energy density of ZBFBs with the Ni/NiO@GF cathode at various current densities. (o) Self-discharge performance comparison of ZBFBs using Ni/NiO@GF and blank GF cathodes.²⁴³ Reproduced with permission from ref. 243. Copyright 2024, Wiley-VCH.

To further optimize the interfacial conductivity and catalytic activity of host materials, researchers have introduced transition metal sulfides to construct heterostructures with enhanced conductivity. This is mainly due to the high electron migration ability of the transition metal–sulfur bonds and the excellent electrocatalytic intrinsic activity brought about by the abundant non-coordinated d orbitals.²⁴⁴ Recently, He et al.²⁴⁵ designed a self-assembled flower-like NiCo₂S₄ nanosheet catalyst and in situ grew it on the surface of GF to construct a NiCo₂S₄-GF composite electrode with confinement and synergistic catalytic properties (Fig. 47a). XPS analysis of the NiCo₂S₄-GF electrode before and after the adsorption of Br₂ reveals its confinement and catalytic effects on bromine species (Fig. 47b). The results show that after Br₂ adsorption, the binding energy of S 2p_3/2 and S 2p_1/2 shifts positively, indicating electron transfer between NiCo₂S₄ and Br₂, which reduces the electron density of sulfur atoms. This indicates effective adsorption of bromine on the host surface, accompanied by interfacial charge redistribution. The active sites created by the synergistic interaction of Ni and Co elements not only stabilize bromine species but also accelerate their redox reaction kinetics, demonstrating the dual role of metal sulfides in halogen confinement and electrocatalytic conversion. The battery constructed with this electrode exhibits a maximum power density of 260.75 mW cm⁻² in polarization testing, demonstrating excellent electrochemical performance. Furthermore, to expand the application of confinement catalysis strategies in halogen systems, Hu et al.²⁴⁶ developed a Co₉S₈@nitrogen-doped carbon (Co₉S₈@NC) catalyst with an inverse perovskite structure (Fig. 47c). This heterostructure achieves tight coupling of cobalt sulfide with nitrogen-doped carbon at the nanoscale and induces charge deficiency characteristics in Co and S through interfacial charge redistribution, which is beneficial for forming stable adsorption with iodine species (I⁻, I⁰, and I⁺) (Fig. 47d). XPS characterization results show that the characteristic peaks of Co and S shift to higher binding energies, while the nitrogen peaks shift to lower binding energies, further confirming the electron transfer behavior from Co₉S₈ to the carbon skeleton (Fig. 47e). This adjustment significantly enhances the host's confinement ability for iodine species and effectively lowers the energy barrier of redox reactions, improving charge transfer rates and electrochemical reaction activity. Structural and performance analysis indicates that Co₉S₈@NC exhibits superior synergy in confinement adsorption and electrocatalytic conversion, promoting reversible conversion of iodine species and effectively suppressing the migration of active materials (Fig. 47f). In situ Raman spectroscopy (Fig. 47g) shows that during the first charge cycle (0.6–1.6 V), the characteristic peaks of I₃⁻ and I₅⁻ gradually increase, indicating the gradual generation of polyiodide anions. In the second stage (1.6–1.8 V), these peaks rapidly decrease, showing that most polyiodide species are further oxidized to I⁺. The subsequent discharge process shows a reversible process, demonstrating good redox reversibility. DFT calculations on the adsorption behaviors of different iodine species (ICl, I₂, I₃⁻, and ZnI₂) on Co₉S₈@NC and a reference NC electrode (Fig. 47h) show that Co₉S₈@NC exhibits lower adsorption energies for iodine species, suggesting superior confinement adsorption, primarily due to its charge-deficient sites’ strong affinity for iodine. Additionally, Gibbs free energy analysis (Fig. 47i) indicates that the key reduction processes on Co₉S₈@NC exhibit negative free energies, suggesting that the reduction processes can occur spontaneously. In particular, the conversion of ICl to ZnI₂ significantly lowers the reaction energy barrier and enhances reaction kinetics. PDOS analysis (Fig. 47j) further reveals strong orbital coupling between Co₉S₈@NC and iodine species, with the d-band centers of Co₉S₈@NC being much closer to the Fermi level (−2.07 and −2.54 eV), compared to NC (−4.32 and −3.71 eV). This indicates stronger electron transfer capabilities and catalytic activity. The corresponding full-cell battery based on Co₉S₈@NC exhibits excellent cycling stability, with a capacity retention rate of 63.21% and high CE at 1 A g⁻¹ and 30 °C (Fig. 47k), outperforming the full cell device with the reference sample NC. Thanks to its high capacity output and stable voltage platform, this bi-electron transfer material system outperforms common zinc-based electrode materials, such as vanadium-based, manganese-based, and Prussian blue materials (Fig. 47l), making it a promising candidate for high-performance Zn–I₂ batteries.

Fig. 47 (a) Schematic of the synthesis process and structure of NiCo₂S₄-GF in Zn–Br₂ batteries. (b) S 2p XPS spectra of NiCo₂S₄-GF and NiCo₂S₄-GF/Br₂ with insets showing the digital photographs of Br₂ solutions containing GF and NiCo₂S₄-GF.²⁴⁵ Reproduced with permission from ref. 245. Copyright 2024, American Chemical Society. (c) SEM image of Co₉S₈@NC. (d) Work function diagram of Co₉S₈@NC. (e) N 1s and Co 2p XPS spectra of Co₉S₈@NC. (f) Mechanism of the interaction between active iodine species and Co₉S₈@NC. (g) In situ Raman spectra of iodine intermediates during charge–discharge processes of the Co₉S₈@NC electrode. (h) Comparison of adsorption energy of ICl, I₂, Zn(I₃)₂, and ZnI₂ on Co₉S₈@NC and NC. (i) Gibbs free energy profiles of Co₉S₈@NC and NC. (j) PDOS plots of I₂ adsorbed on NC and Co₉S₈@NC electrodes. (k) Cycling performance of the Co₉S₈@NC electrode at 1 A g⁻¹. (l) Comparison of energy density, voltage, and capacity of the Co₉S₈@NC electrode with reported zinc-based battery electrodes.²⁴⁶ Reproduced with permission from ref. 246. Copyright 2024, Wiley-VCH.

In summary, heterostructure engineering has proven effective in constructing multiphase structures with interfacial synergistic effects, achieving efficient confinement of halogen species, electrocatalytic conversion, and enhanced electronic/ionic conductivity in zinc–halogen batteries, thus offering a highly promising material design strategy. This approach is centered on the confinement–catalysis–conduction triad: the heterostructures provide abundant interfacial adsorption sites, effectively suppressing the shuttle and dissolution of multivalent halogen species such as I⁻/I₃⁻/I⁺ and Br⁻/Br₂. Simultaneously, the charge transfer behavior and band structure regulation between different phases bestow excellent electrocatalytic activity to the material, accelerating redox kinetics. Moreover, highly conductive carbon-based or metal oxide materials further strengthen electron/ion transport channels, improving rate performance and cycling stability of cathodes. Looking ahead, further research could focus on the following directions: (i) designing interface structures with gradient charge distributions or spontaneous electron migration capabilities to enable precise confinement and stable conversion of halogen intermediates; (ii) exploring multi-component synergistic heterostructures to achieve functional integration and coordinated regulation of “heterogeneous confinement–heterogeneous catalysis–heterogeneous conduction”; and (iii) combining in situ electrochemical spectroscopy with multiscale simulations to deepen the understanding of the intrinsic correlations between electronic redistribution, band regulation, and the redox behavior of multivalent halogen species in heterostructures. These efforts will further accelerate the development of high-performance zinc–halogen batteries.

5. Summary and outlook

Zinc–halogen batteries have become a highly competitive energy storage technology due to their high energy density, environmental friendliness, and excellent safety features. However, they still face challenges such as poor halogen conductivity, high solubility, and the shuttle effect of polyhalides, leading to insufficient reaction kinetics, low CE, and poor cycle life, which limit their further practical applications. Developing halogen-confining host materials with high conductivity, adsorption and catalytic capabilities has been demonstrated to be a highly effective strategy, as reported in recent studies, for optimizing halogen utilization at the cathode and enhancing the reversible conversion of halogen species. Based on the working mechanism of zinc–halogen batteries and related research findings, we summarize and propose a confinement–catalysis–conduction concept for designing host materials and explore the structure–performance relationships and mechanistic insights of various host materials for different types of zinc–halogen batteries. To further improve the reaction kinetics of halogen-confining cathodes, several modification strategies are recommended, including rational structural design, surface functionalization, heteroatom doping, atomic-level engineering of SACs/DACs, and heterostructure engineering. These strategies have proven effective in confinement–catalysis–conduction, leading to improved cycle stability, CE and energy density of zinc–halogen batteries.

Despite substantial advancements in the modification of halogen-confining materials for zinc–halogen batteries, this area of research remains in its early stages, with device performance still failing to meet the requirements for large-scale commercialization. Therefore, it is imperative to explore and develop advanced host materials, innovative synthesis strategies, and cutting-edge characterization techniques to address the existing challenges. These efforts are crucial for enhancing the competitiveness of zinc–halogen batteries in the energy storage sector and facilitating their large-scale practical deployment. Based on the findings of this review, the following suggestions are proposed (Fig. 48).

Fig. 48 Future development prospects of zinc–halogen batteries.

5.1. Mechanism-oriented structural design paradigm for halogen-confining host materials

With the continuous development of zinc–halogen batteries, the research focus has gradually shifted from empirically driven material screening to a more systematic approach to host material design, particularly under the guidance of the confinement–catalysis–conduction mechanism for innovative approaches. Future research is suggested to explore several novel design strategies: first, the construction of host materials with ordered hierarchical channels should be pursued to enable multi-scale pore structure synergy. This approach would effectively confine halogen species at the microscopic scale while ensuring electrolyte permeability at the macroscopic scale, thus improving the utilization efficiency of active materials and alleviating diffusion polarization issues. Second, anchoring highly dispersed multi-atom catalysts on conductive frameworks as host materials or directly constructing conductive conjugated complexes (e.g., NiPPc) with analogous functionalities is recommended,²⁸ which integrates the advantages of physical confinement, chemical adsorption, and electrocatalytic conversion,²²⁶ thereby enhancing the efficient electrocatalytic conversion of halogen species and optimizing redox kinetics. Third, heteroatom doping and functional modification of host materials can be used to adjust the electronic structure and adsorption selectivity, enhancing their stability and selective recognition capability toward specific halogen intermediates. Additionally, inspired by biological recognition mechanisms, responsive host–guest networks such as cyclodextrins or starch helical cavities can be constructed as host materials to facilitate the reversible release and dynamic confinement of multi-halide ions, which would improve the structural stability of the materials and extend the cycling life of the batteries. Lastly, to meet the requirements for high energy density, it is necessary to increase halogen loading. For example, the iodine content and iodine loading in the cathode are typically below 50% and 3 mg cm⁻², respectively.²⁴⁷ Therefore, there is an urgent need to investigate and develop host materials with high halogen loading capacity and stable confinement–catalysis–conduction functions. These strategies are expected to address the critical challenges faced by zinc–halogen batteries, leading to significant improvements in battery performance.

5.2. Interdisciplinary operando techniques for multiscale mechanistic insights

Traditional ex situ or single characterization techniques are inadequate for capturing the complex structural changes in real-time, hindering a deeper understanding of the structure–performance–mechanism relationship. In contrast, in situ characterization techniques offer dynamic, real-time, and direct insights, allowing for precise monitoring of structure, valence, and phase changes occurring within the battery under operating conditions. Therefore, constructing a multidimensional operando analysis approach is crucial for the understanding of the underlying mechanisms. At the atomic and molecular scales, DFT calculations can be used to analyze the adsorption configurations, migration kinetics, and electronic structural characteristics of halogen species (e.g., I₃⁻ and Br₃⁻) at different catalytic sites. The minimum energy path method can be applied to predict the energy barriers of key reaction steps, while Ab initio molecular dynamics (AIMD) simulations can model the dynamic evolution and interface interactions at finite temperatures. For example, Yang et al.¹⁹¹ utilized DFT and AIMD to reveal the energy barrier matching advantages of Cu SACs in the dissociation–conversion of I₃⁻, and through d-band center analysis, they enhanced charge transfer efficiency and reaction reversibility. In macro-scale reactions, integrating various in situ techniques (such as UV-vis, Raman, XPS, NMR, and synchrotron XAS) allows for real-time or near-real-time monitoring of halogen intermediate concentration changes, electrode–electrolyte interface reaction pathways, and the evolution of metal center oxidation states and coordination structures. For example, operando Mössbauer spectroscopy has been successfully applied in the lithium–iron system to dynamically monitor the Fe²⁺/Fe³⁺ redox state conversion,²⁴⁸ demonstrating its high sensitivity in tracking transition metal active sites. This technique could potentially be applied to investigate the redox behavior of halogen species. In terms of visualization, low-temperature imaging techniques such as cryo-TEM and cryo-ET can freeze the electrode structure, capturing the spatial morphology and structural features of confinement cavities, providing direct evidence for the localized enrichment and diffusion pathways of halogen intermediates. Scanning electrochemical microscopy employs ultramicroelectrodes to probe the spatially resolved current distribution, electrochemical activity, and mass transport behavior at localized regions of the electrode surface. Therefore, the construction of a multi-scale operando characterization system, focusing on electronic structure regulation, interface behavior monitoring, and confinement behavior visualization, is expected to drive the shift of host material design from empirical-driven to mechanism-guided approaches.

5.3. AI-assisted screening and precise structural tuning of host materials

AI technology is profoundly reshaping the traditional paradigm of material development. In material screening, previous studies have constructed databases incorporating material structural parameters, local electronic state characteristics, adsorption energies, pore size distributions, and conductivity through high-throughput first-principles calculations, combined with ML models such as GNN, generative adversarial networks, random forests, support vector machines, and Bayesian optimization. These models enable the intelligent screening of host materials and the efficient modeling of structure–property relationships. For example, Zhou et al.²⁴⁹ used a crystal graph convolutional neural network in conjunction with the AFLOW database to screen approximately 80 predicted Zn-ion battery cathode materials from over 130 000 inorganic materials, about 10 of which were experimentally verified, showing that the predictions were highly consistent with experimental results. This model can also be used to efficiently screen host materials with good halogen species confinement capability, adsorption capacity, and electrical conductivity. In terms of structural design, AI-assisted screening has revealed various strategies for tuning host materials, such as micro-to-macro pore synergistic construction, introduction of single-atom catalytic sites, modification with halogen-affinity functional groups, and interface polarity regulation, thereby enhancing the materials’ adsorption selectivity and catalytic efficiency. Therefore, the integration of AI-driven high-throughput predictions with ML models holds great promise for accelerating the screening and precise design of host materials for zinc–halogen batteries.

5.4 Synergistic regulation of interfacial behaviors among cathodes, anodes and electrolytes

In the zinc–halogen batteries, the stability and dynamic evolution of the cathode electrode/electrolyte interface directly impact the adsorption, conversion, and release efficiency of halogen species, making it a key factor determining capacity retention, self-discharge rate, and rate performance. Halogen intermediates are prone to dissolution-migration-re-deposition processes during charge and discharge, leading to shuttle effects and reaction irreversibility. Therefore, constructing a cathode-electrolyte functionalized interface with synergistic confinement and electrocatalytic capabilities can achieve highly selective adsorption, interface charge transfer regulation, and reaction pathway guidance. Meanwhile, the behavior at the cathode–anode interface affects polarization and dendrite formation. During battery operation, the halogen conversion behavior at the positive electrode is closely related to the reversible deposition-stripping process of Zn²⁺ at the negative electrode. If there are significant differences in the reaction kinetics of the two, it can lead to increased polarization and dendrite growth, affecting the cycle stability. By adjusting the surface polarity and electron density distribution of the positive electrode, the concentration gradient of Zn²⁺ and halide ions in the electrolyte can be controlled, thereby stabilizing the negative electrode interface reaction. Alternatively, a dual-functional ion-selective interface layer (e.g., Janus interface) can be used to precisely regulate charge distribution and ion migration,²⁵⁰ alleviating inhomogeneity. Moreover, controlling the Zn anode/electrolyte interface is also crucial. Constructing interface stabilizing layers (e.g., ZnF₂)²⁵¹ or introducing asymmetric anion regulators (e.g., TFSI⁻ and OTf⁻)²⁹ or co-solvents (e.g., ethylene glycol)²⁹ to regulate the electrolyte solvation structure helps suppress side reactions and dendrite growth, improving interface charge distribution and enhancing the cycle stability and CE of batteries. Therefore, synergistic regulation involving the cathode–electrolyte interface, electrode matching optimization, and stable construction of the Zn anode interface is essential for achieving high-performance zinc–halogen batteries.

5.5. Scalability and engineering challenges of zinc–halogen batteries

Zinc–halogen batteries have broad prospects in large-scale energy storage, but their commercialization process is still constrained by host material selection, manufacturing costs, and system stability. For example, carbon-based materials have high specific surface area, tunable pore structure, and excellent conductivity, which help to suppress halogen species diffusion and enhance cycle life, while their low cost (15–50 USD/kg) makes them suitable for industrial applications.²⁵² However, their wide pore size distribution and surface chemical defects limit their ability to adsorb halogen anions, which poses challenges for scale-up. Functionalized framework materials (such as MOFs, COFs, etc.) offer designable pore channels and chemical anchoring sites for efficient halogen species adsorption, but they are costly (e.g., Ni-MOF-74 ≈ 886 USD per kg) and have poor cycle life, facing issues such as high energy consumption and safety control during scaling.²⁵³ Carbon derived from framework materials exhibits excellent performance due to atomic-level active sites and tunable electronic structures, but its high synthesis cost, complex process, and lack of consistency and controllability remain significant hurdles.²⁵⁴ 2D transition metal carbides/nitrides stand out due to their good conductivity and cycle stability, but their synthesis methods still rely on high-energy consumption, hazardous processes, and low yields, resulting in high material costs (>10 000 USD per kg),²⁵⁵ and there is no mature, green, and efficient large-scale synthesis method. Topological insulator materials theoretically have high selective anchoring capabilities for halogen species and can suppress shuttle effects, but their synthesis methods and electrochemical performance studies are still in the early stages. Additionally, most of the excellent electrochemical performance data for zinc–halogen batteries are derived from laboratory-prepared cells evaluated under ideal testing conditions, such as low areal active material loading (≈2 mg cm⁻²).²⁵⁶ However, for industrial-scale production, evaluation criteria must consider several factors, including higher areal active material loading (>10 mg cm⁻²), limited electrolyte volume, and optimized anode/cathode capacity ratios. The inconsistency in testing standards significantly hinders the commercialization of zinc–halogen batteries and requires urgent resolution. Notably, in 2023, Zn–Br₂ flow batteries made initial progress at the industrial level, with TETRA PureFlow ultra-pure zinc bromide fluid being utilized for grid energy storage and photovoltaic systems.²⁵⁷ In 2024, Jiangsu Heng’an Energy Storage Technology Co., Ltd established a production facility for Zn–Br₂ flow batteries with an annual capacity of 10 GWh.²⁵⁸ Nevertheless, the transition from laboratory-scale to large-scale applications of zinc–halogen batteries still faces challenges such as poor coating consistency, significant efficiency losses, and high voltage drops in battery stacks, which urgently need to be addressed.

Author contributions

Shude Liu conceived and supervised the overall project, made substantial contributions to the conceptualization, writing, and critical revision of the manuscript, and coordinated all stages of its development. Xue Peng, Yafei Chai, Ming Ma and Ling Kang conducted the literature survey and prepared the figures under the guidance of Shude Liu. Huilin Zhang and Jieming Chen supported the literature compilation and assisted in figure visualization. Bin Ding, Yusuke Yamauchi, and Seong Chan Jun critically reviewed the manuscript and provided valuable suggestions to enhance its clarity, structure, and scientific depth. All authors contributed to the discussion of the content and approved the final version of the manuscript.

Conflicts of interest

The authors declare no competing financial interests.

Data availability

This review does not report any new datasets, software, or code. All information presented is based on previously published sources, which are properly cited throughout the manuscript.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 52302223 and 52302251), the Fundamental Research Funds for the Central Universities (2232024G-06-01), the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (RS-2024-00339770), Korea Environment Industry & Technology Institute (KEITI) through Technology Development Project for Biological Hazards Management in Indoor Air Program (or Project) funded by Korea Ministry of Environment (MOE) (ARQ202101038001), the Science and Technology Commission of Shanghai Municipality (24xtcx00300), the National Natural Science Foundation Joint Fund for Regional Innovation and Development Fund (U24A2073), the ARC Laureate Fellowship (FL230100095), and the JST-ERATO Yamauchi Materials Space-Tectonics Project (JPMJER2003).

Notes and references

1. Y. Guan, J. Yan, Y. Shan, Y. Zhou, Y. Hang, R. Li, Y. Liu, B. Liu, Q. Nie and B. Bruckner, Nat. Energy, 2023, 8, 304–316.
2. R. Liu, G. He, X. Wang, D. Mallapragada, H. Zhao, Y. Shao-Horn and B. Jiang, Nat. Commun., 2024, 15, 280.
3. F. Degen, M. Winter, D. Bendig and J. Tübke, Nat. Energy, 2023, 8, 1284–1295.
4. Y. Meng, D. Zhou, R. Liu, Y. Tian, Y. Gao, Y. Wang, B. Sun, F. Kang, M. Armand and B. Li, Nat. Energy, 2023, 8, 1023–1033.
5. W. Ma, S. Wang, X. Wu, W. Liu, F. Yang, S. Liu, S. C. Jun, L. Dai, Z. He and Q. Zhang, Energy Environ. Sci., 2024, 17, 4819–4846.
6. S. Ou, J. Zheng, X. Chen, R. Li, Z. Yuan, S. Liu, Y. Niu, M. An, G. Zhou and Y. Yamauchi, Energy Environ. Sci., 2025, 18, 5457–5469.
7. Y. Liang and Y. Yao, Nat. Rev. Mater., 2023, 8, 109–122.
8. L. Jiang, Z. Xing, Y. Liu, X. Shi, L. Li, Y. Liu, B. Lu and J. Zhou, Energy Environ. Sci., 2025, 18, 3005–3013.
9. S. Li, Y. Zhong, J. Huang, G. Lai, L. Li, L. Jiang, X. Xu, B. Lu, Y. Liu and J. Zhou, Energy Environ. Sci., 2025, 18, 2599–2609.
10. H. Chen, W. Zhang, S. Yi, Z. Su, Z. Zhao, Y. Zhang, B. Niu and D. Long, Energy Environ. Sci., 2024, 17, 3146–3156.
11. G. Qu, H. Wei, S. Zhao, Y. Yang, X. Zhang, G. Chen, Z. Liu, H. Li and C. Han, Adv. Mater., 2024, 36, 2400370.
12. M. Chen, Y. Gong, Y. Zhao, Y. Song, Y. Tang, Z. Zeng, S. Liang, P. Zhou, B. Lu and X. Zhang, Natl. Sci. Rev., 2024, 11, nwae205.
13. C. Li, Q. Li, Z. Wu, Y. Wang, R. Zhang, H. Cui, Y. Hou, J. Liu, Z. Huang and C. Zhi, Adv. Mater., 2024, 36, 2304878.
14. N. J. Herrmann, H. Euchner, A. Groß and B. Horstmann, Adv. Energy Mater., 2024, 14, 2302553.
15. T. Wang, S. Li, X. Weng, L. Gao, Y. Yan, N. Zhang, X. Qu, L. Jiao and Y. Liu, Adv. Energy Mater., 2023, 13, 2204358.
16. R. Kushwaha, C. Jain, P. Shekhar, D. Rase, R. Illathvalappil, D. Mekan, A. Camellus, C. P. Vinod and R. Vaidhyanathan, Adv. Energy Mater., 2023, 13, 2301049.
17. S. J. Zhang, J. Hao, H. Wu, Q. Chen, C. Ye and S. Z. Qiao, Adv. Mater., 2024, 36, 2404011.
18. T. Wang, P. Wang, L. Pan, Z. He, L. Dai, L. Wang, S. Liu, S. C. Jun, B. Lu and S. Liang, Adv. Energy Mater., 2023, 13, 2203523.
19. X. Zhao, J. Hao, Q. Chen, S. J. Zhang, H. Wu, L. Mao and S. Z. Qiao, Angew. Chem., Int. Ed., 2025, 64, e202502386.
20. S. Lv, T. Fang, Z. Ding, Y. Wang, H. Jiang, C. Wei, D. Zhou, X. Tang and X. Liu, ACS Nano, 2022, 16, 20389–20399.
21. C. Wang, Z. Lv, W. Yang, X. Feng and B. Wang, Chem. Soc. Rev., 2023, 52, 1382–1427.
22. X. Li, W. Xu and C. Zhi, Nat. Rev. Chem., 2024, 8, 359–375.
23. S. Wang, Y. Wang, Z. Wei, J. Zhu, Z. Chen, H. Hong, Q. Xiong, D. Zhang, S. Li and S. Wang, Adv. Mater., 2024, 36, 2401924.
24. P. Li, S. Yang, J. Zhu, S. Wang, Y. Hou, H. Cui, Z. Chen, R. Zhang, Z. Wu and Y. Wang, Matter, 2024, 7, 1867–1878.
25. L. Hu, C. Dai, Y. Zhu, X. Hou, Z. Liu, X. Geng, H. Wang, J. Chen, N. Sun and Q. Rong, Energy Environ. Sci., 2024, 17, 5552–5562.
26. R. Wang, Z. Liu, J. Wan, X. Zhang, D. Xu, W. Pan, L. Zhang, H. Li, C. Zhang and Q. Zhang, Adv. Energy Mater., 2024, 14, 2402900.
27. W. Li and D. Wang, Adv. Mater., 2025, 37, 2304983.
28. H. Wei, G. Qu, X. Zhang, B. Ren, S. Li, J. Jiang, Y. Yang, J. Yang, L. Zhao and H. Li, Energy Environ. Sci., 2023, 16, 4073–4083.
29. H. Xu, R. Zhang, D. Luo, K. Huang, J. Wang, G. Sun, H. Dou and X. Zhang, Energy Environ. Sci., 2025, 18, 7447–7459.
30. H. Huang, Y. Fan, Y. Wang, L. Wang, Y. Jiang, Y. Cheng, J. Wang, Y. Zhu and Y. Yang, Matter, 2025, 8, 102154.
31. M. Liu, Q. Chen, X. Cao, D. Tan, J. Ma and J. Zhang, J. Am. Chem. Soc., 2022, 144, 21683–21691.
32. J.-M. Cao, I. V. Zatovsky, Z.-Y. Gu, J.-L. Yang, X.-X. Zhao, J.-Z. Guo, H. Xu and X.-L. Wu, Prog. Mater. Sci., 2023, 135, 101105.
33. X. Guo, H. Xu, Y. Tang, Z. Yang, F. Dou, W. Li, Q. Li and H. Pang, Adv. Mater., 2024, 36, 2408317.
34. Y. Zhang, C. Wei, M.-X. Wu, Y. Wang, H. Jiang, G. Zhou, X. Tang and X. Liu, Chem. Eng. J., 2023, 451, 138915.
35. Y. Shu, D. Wang, J. Wang and H. Huang, Chem. Eng. J., 2024, 498, 154925.
36. X. Tang, X. Guo, W. Wu and G. Wang, Adv. Energy Mater., 2018, 8, 1801897.
37. L. She, H. Cheng, Z. Yuan, Z. Shen, Q. Wu, W. Zhong, S. Zhang, B. Zhang, C. Liu and M. Zhang, Adv. Sci., 2024, 11, 2305061.
38. H. Chen, X. Li, K. Fang, H. Wang, J. Ning and Y. Hu, Adv. Energy Mater., 2023, 13, 2302187.
39. H. Wen, H. Wen, Y. Sun, J. Li, C. Zhuang, M. Liu, H. Li, H. Fan, B. Yin and S. Zhang, Adv. Funct. Mater., 2025, 35, 2500323.
40. Z. Yan, Q. Yang and C. Yang, J. Mater. Chem. A, 2024, 12, 24746–24760.
41. X. Chen, X. Xie, P. Ruan, S. Liang, W.-Y. Wong and G. Fang, ACS Energy Lett., 2024, 9, 2037–2056.
42. J. H. Lee, Y. Byun, G. H. Jeong, C. Choi, J. Kwen, R. Kim, I. H. Kim, S. O. Kim and H. T. Kim, Adv. Mater., 2019, 31, 1904524.
43. C. Wang, W. Lu, Q. Lai, P. Xu, H. Zhang and X. Li, Adv. Mater., 2019, 31, 1904690.
44. X. Yuan, J. Mo, J. Huang, J. Liu, C. Liu, X. Zeng, W. Zhou, J. Yue, X. Wu and Y. Wu, ChemElectroChem, 2020, 7, 1531–1536.
45. D. Yu, A. Kumar, T. A. Nguyen, M. T. Nazir and G. Yasin, ACS Sustainable Chem. Eng., 2020, 8, 13769–13776.
46. L. Ma, Y. Ying, S. Chen, Z. Huang, X. Li, H. Huang and C. Zhi, Angew. Chem., Int. Ed., 2021, 133, 3791–3798.
47. W. Lu, P. Xu, S. Shao, T. Li, H. Zhang and X. Li, Adv. Funct. Mater., 2021, 31, 2102913.
48. P. Xu, T. Li, Q. Zheng, H. Zhang, Y. Yin and X. Li, J. Energy Chem., 2022, 65, 89–93.
49. W. Gao, S. Cheng, Y. Zhang, E. Xie and J. Fu, Adv. Funct. Mater., 2023, 33, 2211979.
50. S. Wang, Z. Huang, B. Tang, X. Li, X. Zhao, Z. Chen, C. Zhi and A. L. Rogach, Adv. Energy Mater., 2023, 13, 2300922.
51. L. Tang, T. Li, W. Lu and X. Li, Energy Environ. Sci., 2024, 17, 3136–3145.
52. N. Chen, W. Wang, Y. Ma, M. Chuai, X. Zheng, M. Wang, Y. Xu, Y. Yuan, J. Sun and K. Li, Small Methods, 2024, 8, 2201553.
53. J. Hu, Z. Zhang, T. Deng, F. C. Cui, X. Shi, Y. Tian and G. Zhu, Adv. Mater., 2024, 36, 2401091.
54. D. Lin and Y. Li, Adv. Mater., 2022, 34, 2108856.
55. Z. Khan, D. Kumar and X. Crispin, Adv. Mater., 2023, 35, 2300369.
56. T. Xiao, J.-L. Yang, D. Chao and H. J. Fan, Natl. Sci. Rev., 2025, 12, nwaf029.
57. Q. Li, K. K. Abdalla, J. Xiong, Z. Song, Y. Wang, Y. Zhao, M. Liu, Y. Fan, Y. Zhao and X.-M. Sun, Energy Mater., 2024, 4, 400040.
58. C. Xie, C. Wang, Y. Xu, T. Li, Q. Fu and X. Li, Nat. Energy, 2024, 9, 714–724.
59. J. Guo, L. Hu, R. Wang, G. Liu, J. Zhao, C. Dai and Z. Lin, ACS Nano, 2025, 19, 9340–9350.
60. Z. Feng, Y. Tang, Y. Wei, J. He, G. Liu, J. Yan, J. Qi, Z. Shi, Q. Yang and Z. Wen, Angew. Chem., Int. Ed., 2025, 137, e202416755.
61. B. He, F. Zhang, Y. Xin, C. Xu, X. Hu, X. Wu, Y. Yang and H. Tian, Nat. Rev. Chem., 2023, 7, 826–842.
62. Y. Wu, M. Chen, G. Zhao, D. Qi, X. Zhang, Y. Li, Y. Huang and W. Yang, Adv. Mater., 2024, 36, 2311717.
63. R. Chen, ChemSusChem, 2025, 18, e202401678.
64. J. Hao, X. Li, X. Zeng, D. Li, J. Mao and Z. Guo, Energy Environ. Sci., 2020, 13, 3917–3949.
65. C. Tang, X. Wang, M. Ma, Z. Wang, Y. Li, H. Li, B. Li, Y. Zhang and X. Zhu, Chem. Eng. J., 2023, 471, 144784.
66. J. Hao, S. Zhang, H. Wu, L. Yuan, K. Davey and S.-Z. Qiao, Chem. Soc. Rev., 2024, 53, 4312–4332.
67. T. A. Shifa and A. Vomiero, Adv. Energy Mater., 2019, 9, 1902307.
68. J.-Y. Chen, Y. Xiao, F.-S. Guo, K.-M. Li, Y.-B. Huang and Q. Lu, ACS Catal., 2024, 14, 5198–5226.
69. Z. Li, H. Yun, Y. Yan, Y. Zhao and F. Zhao, Angew. Chem., Int. Ed., 2025, 64, e202506560.
70. D. Xie, Z. W. Wang, Z. Y. Gu, W. Y. Diao, F. Y. Tao, C. Liu, H. Z. Sun, X. L. Wu, J. W. Wang and J. P. Zhang, Adv. Funct. Mater., 2022, 32, 2204066.
71. K. Zhang and Z. Jin, Energy Storage Mater., 2022, 45, 332–369.
72. D. Pan, Y. Fu, J. Chen, K. J. Czech, J. C. Wright and S. Jin, Nano Lett., 2018, 18, 1807–1813.
73. Z. Xue, Z. Gao and X. Zhao, Energy Environ. Mater., 2022, 5, 1155–1179.
74. P. De, J. Halder, C. C. Gowda, S. Kansal, S. Priya, S. Anshu, A. Chowdhury, D. Mandal, S. Biswas and B. K. Dubey, Electrochem. Sci. Adv., 2023, 3, e2100159.
75. S. Chatterji, Cem. Concr. Res., 1995, 25, 299–303.
76. J. Chen, X. Song, R. Jiang, F. Yang, Y. Li, L. Liao, Y. Shen, B. Guo, B. Yu and Z. Ma, Chem. Eng. J., 2024, 481, 148332.
77. Y. Jiang, W. Li, X. Li, Y. Liao, X. Liu, J. Yu, S. Xia, W. Li, B. Zhao and J. Zhang, J. Colloid Interface Sci., 2024, 672, 287–298.
78. S. M. Scott, T. Hu, T. Yao, G. Xin and J. Lian, Carbon, 2015, 90, 1–8.
79. J. Xu, W. Ma, L. Ge, M. Ren, X. Cai, W. Liu, J. Yao, C. Zhang and H. Zhao, J. Alloys Compd., 2022, 912, 165151.
80. J. Sun, H. Ma and D. Wang, J. Alloys Compd., 2023, 947, 169696.
81. W. Xie, D. Cui, S.-R. Zhang, Y.-H. Xu and D.-L. Jiang, Mater. Horiz., 2019, 6, 1571–1595.
82. Y. Xie, T. Pan, Q. Lei, C. Chen, X. Dong, Y. Yuan, W. A. Maksoud, L. Zhao, L. Cavallo and I. Pinnau, Nat. Commun., 2022, 13, 2878.
83. Z. Zhu, D. Luo, L. Wei, J. Chen, T. Jia, X. Chi and X. Guo, Energy Storage Mater., 2025, 75, 103994.
84. Y. Qian, G. Chang, C. Huang, Y. Yang, Y. Hao, Q. Tang, A. Hu, Y. Li and X. Chen, Chem. Eng. J., 2025, 503, 158392.
85. A. Mohan, M. H. Al-Sayah, A. Ahmed and O. M. El-Kadri, Sci. Rep., 2022, 12, 2638.
86. D. Li, Y. J. Zhu, L. Cheng, S. Xie, H. P. Yu, W. Zhang, Z. Xu, M. G. Ma and H. Li, Adv. Energy Mater., 2025, 15, 2404426.
87. W. Du, Q. Huang, X. Zheng, Y. Lv, L. Miao, Z. Song, L. Gan and M. Liu, Energy Environ. Sci., 2025, 18, 6540–6547.
88. Y. Zhai, Y. Dou, D. Zhao, P. F. Fulvio, R. T. Mayes and S. Dai, Adv. Mater., 2011, 23, 4828–4850.
89. J. Xu, Z. Huang, H. Zhou, G. He, Y. Zhao and H. Li, Energy Storage Mater., 2024, 72, 103596.
90. J. Chen, H. Zhang, X. Peng, X. Shao, Y. Chai, M. Ma, Z. Li, S. Liu and B. Ding, Rare Met., 2025 DOI:10.1007/s12598-025-03484-8.
91. H. Pan, B. Li, D. Mei, Z. Nie, Y. Shao, G. Li, X. S. Li, K. S. Han, K. T. Mueller and V. Sprenkle, ACS Energy Lett., 2017, 2, 2674–2680.
92. H. Yu, X. Cai, Z. Wang, Z. Yang, W. Liu, M. Ren, J. Yao, Q. Liu and C. Qiao, Colloids Surf., A, 2024, 684, 133239.
93. Y. X. Gan, C, 2021, 7, 39.
94. Y. Wang, M. Zhang, X. Shen, H. Wang, H. Wang, K. Xia, Z. Yin and Y. Zhang, Small, 2021, 17, 2008079.
95. X. Feng, Y. Bai, M. Liu, Y. Li, H. Yang, X. Wang and C. Wu, Energy Environ. Sci., 2021, 14, 2036–2089.
96. L. Wei, X. Zhang, X. Li, C. Chen, D. Yu and G. Zhao, J. Electroanal. Chem., 2024, 967, 118452.
97. J. Ma, A. Azizi, E. Zhang, H. Zhang, A. Pan and K. Lu, Chem. Sci., 2024, 15, 4581–4589.
98. Q. Chen, S. Chen, J. Ma, S. Ding and J. Zhang, Nano Energy, 2023, 117, 108897.
99. M. K. Wong, J. J. Foo, J. Y. Loh and W. J. Ong, Adv. Energy Mater., 2024, 14, 2303281.
100. J. Yang, Y. Kang, F. Meng, W. Meng, G. Chen, M. Zhang, Z. Lv, Z. Wen, C. C. Li and J. Zhao, Energy Environ. Sci., 2025, 18, 236–245.
101. X. Zhang, J. Maddock, T. M. Nenoff, M. A. Denecke, S. Yang and M. Schröder, Chem. Soc. Rev., 2022, 51, 3243–3262.
102. F. Wei, T. Zhang, H. Xu, Y. Peng, H. Guo, Y. Wang, S. Guan, J. Fu, C. Jing and J. Cheng, Adv. Funct. Mater., 2024, 34, 2310693.
103. C. Li, Y. Xiong, X. Shen, D. Luo, W. Wu, Z. Li, M. H. Helal, Z. M. El-Bahy, Y. Mai and Z. Liu, Coord. Chem. Rev., 2024, 520, 216100.
104. Y. Cui, B. Li, H. He, W. Zhou, B. Chen and G. Qian, Acc. Chem. Res., 2016, 49, 483–493.
105. M. Wang, R. Dong and X. Feng, Chem. Soc. Rev., 2021, 50, 2764–2793.
106. P. Bao, L. Cheng, X. Yan, X. Nie, X. Su, H. G. Wang and L. Chen, Angew. Chem., Int. Ed., 2024, 136, e202405168.
107. J. He, Y. Mu, B. Wu, F. Wu, R. Liao, H. Li, T. Zhao and L. Zeng, Energy Environ. Sci., 2024, 17, 323–331.
108. Y. Liu, D. Liu, Z. Liu, M. Mao, D. Zhang, J. Tao, H. Zhang, K. Song, L. Liu and Y. Han, J. Am. Chem. Soc., 2025, 147, 3959–3966.
109. L. Yin, X. Guo, J. Hu, K. Yan, L. Liu, X. Shi, F. Cui, G. Zhu and N. Zhang, Angew. Chem., Int. Ed., 2025, 64, e202423265.
110. K. Geng, T. He, R. Liu, S. Dalapati, K. T. Tan, Z. Li, S. Tao, Y. Gong, Q. Jiang and D. Jiang, Chem. Rev., 2020, 120, 8814–8933.
111. W. Li, L. Huang, H. Zhang, Y. Wu, F. Wei, T. Zhang, J. Fu, C. Jing, J. Cheng and S. Liu, Matter, 2023, 6, 2312–2323.
112. L. Huang, W. Li, F. Wei, S. Ke, H. Chen, C. Jing, J. Cheng and S. Liu, Chemistry, 2024, 10, 3100–3113.
113. J.-D. Feng, W.-K. Han, J.-J. He, Y. Liu, R.-M. Zhu, J. Zhang, H. Pang and Z.-G. Gu, Angew. Chem., Int. Ed., 2025, 64, e202506006.
114. H. Wang, Y. Jin, N. Sun, W. Zhang and J. Jiang, Chem. Soc. Rev., 2021, 50, 8874–8886.
115. D. Huang, X. Lu, K. Zhang, Y. Wang, F. Wang, H. Gao, Y. Ding, L. Pan, Y. Li and Y. Li, Chem. Eng. J., 2024, 494, 153256.
116. L. Zhang, K. Luo, J. Gong, Y. Zhou, H. Guo, Y. Yu, G. He, J.-F. Gohy, I. P. Parkin and J. Hofkens, Angew. Chem., Int. Ed., 2025, 64, e202506822.
117. C. Xu, Z. Yang, X. Zhang, M. Xia, H. Yan, J. Li, H. Yu, L. Zhang and J. Shu, Nano-Micro Lett., 2021, 13, 1–36.
118. L. Ma, Y. Ying, S. Chen, Z. Huang, X. Li, H. Huang and C. Zhi, Angew. Chem., Int. Ed., 2021, 60, 3791–3798.
119. H. Ali, Y. Orooji, A. Y. A. Alzahrani, H. M. Hassan, Z. Ajmal, D. Yue and A. Hayat, ACS Nano, 2025, 19, 7482–7545.
120. Y. Zhao, W. Xue, Z. Jiang, D. Liu, C. Hu and H. Huang, ACS Sustainable Chem. Eng., 2023, 11, 6741–6751.
121. Z. Wang, X. Zou, M. Lv and B. Zhang, Energy Mater., 2024, 4, 400072.
122. Y. Zhao, Y. Wang, W. Xue, R. Cheng, X. Zheng, G. Zhu, D. Hu, H. Huang, C. Hu and D. Liu, Adv. Mater., 2024, 36, 2403097.
123. G. Cai, P. Yan, L. Zhang, H.-C. Zhou and H.-L. Jiang, Chem. Rev., 2021, 121, 12278–12326.
124. Z. Chen, Y. Guo, L. Han, J. Zhang, Y. Liu, J. Baeyens and Y. Lv, Prog. Energy Combust. Sci., 2024, 104, 101175.
125. L. Chai, X. Wang, Y. Hu, X. Li, S. Huang, J. Pan, J. Qian and X. Sun, Adv. Sci., 2022, 9, 2105063.
126. J. Ren, Y. Huang, H. Zhu, B. Zhang, H. Zhu, S. Shen, G. Tan, F. Wu, H. He and S. Lan, Carbon Energy, 2020, 2, 176–202.
127. L. Wang, T. Li, X. Dong, M. Pang, S. Xiao and W. Zhang, Chem. Eng. J., 2021, 425, 130578.
128. N. Li, Z. Yang, Y. Li, D. Yu, T. Pan, Y. Chen, W. Li, H. Xu, X. Guo and H. Pang, Adv. Energy Mater., 2024, 14, 2402846.
129. H. Zhang, S. Wang, E. Lv, C. He, X.-L. Dong, J. Qiu, Y. Wang and Z. Wen, Energy Environ. Sci., 2025, 18, 3986–4009.
130. X. Yang, H. Fan, F. Hu, S. Chen, K. Yan and L. Ma, Nano-Micro Lett., 2023, 15, 126.
131. M. Risch, F. Ringleb, M. Kohlhoff, P. Bogdanoff, P. Chernev, I. Zaharieva and H. Dau, Energy Environ. Sci., 2015, 8, 661–674.
132. W. Han, J. Zhao and X. Li, J. Alloys Compd., 2024, 1008, 176647.
133. X. Li, N. Li, Z. Huang, Z. Chen, G. Liang, Q. Yang, M. Li, Y. Zhao, L. Ma and B. Dong, Adv. Mater., 2021, 33, 2006897.
134. C. Shi, X. Zhang, Z. Li, T. T. Beyene, T. Zheng, Y. Liu and K. Zhu, Energy Fuels, 2024, 38, 14866–14890.
135. W. Yan, Y. Liu, J. Qiu, F. Tan, J. Liang, X. Cai, C. Dai, J. Zhao and Z. Lin, Nat. Commun., 2024, 15, 9702.
136. F. Liu and Z. Fan, Chem. Soc. Rev., 2023, 52, 1723–1772.
137. Z. Zhang, W. Ling, N. Ma, J. Wang, X. Chen, J. Fan, M. Yu and Y. Huang, Adv. Funct. Mater., 2024, 34, 2310294.
138. S. J. Zhang, J. Hao, H. Li, P. F. Zhang, Z. W. Yin, Y. Y. Li, B. Zhang, Z. Lin and S. Z. Qiao, Adv. Mater., 2022, 34, 2201716.
139. L. Zhang, M. Zhang, H. Guo, Z. Tian, L. Ge, G. He, J. Huang, J. Wang, T. Liu and I. P. Parkin, Adv. Sci., 2022, 9, 2105598.
140. B. Qi, Y. Chai, Y. Hu, Z. Liu, Y. Wang, K. Chen, C. Tan, X. Bai, K. Xie and H. Cheng, Angew. Chem., Int. Ed., 2025, 64, e202507497.
141. P. Ding, B. Chang, G. Qing and T. Sun, Sci. China Chem., 2014, 57, 1492–1506.
142. R. Zhang, C. Hong, C. Yu and Y. Zhou, Chem. – Eur. J., 2025, 31, e202500763.
143. J. Chen, G. Adit, L. Li, Y. Zhang, D. H. Chua and P. S. Lee, Energy Environ. Mater., 2023, 6, e12633.
144. H. Xu, W. Gao, H. Dou, X. Zhang and M. Pumera, Adv. Funct. Mater., 2025, 35, 2414022.
145. E. Ballas, A. Nimkar, G. Bergman, R. Elazari, R. Wakshlak, D. Sharon, M. D. Levi, D. T. Major, D. Mandler and N. Shpigel, Energy Storage Mater., 2024, 70, 103461.
146. D. Han and S. Shanmugam, Curr. Opinion Electrochem., 2024, 45, 101485.
147. A. Mahmood, Z. Zheng and Y. Chen, Adv. Sci., 2024, 11, 2305561.
148. C. Wang, X. Li, X. Xi, P. Xu, Q. Lai and H. Zhang, RSC Adv., 2016, 6, 40169–40174.
149. C. Wang, Q. Lai, P. Xu, D. Zheng, X. Li and H. Zhang, Adv. Mater., 2017, 29, 1605815.
150. Z. Li, X. Wu, X. Yu, S. Zhou, Y. Qiao, H. Zhou and S.-G. Sun, Nano Lett., 2022, 22, 2538–2546.
151. Y. Li, S. Wang, Z. Lv, Z. Wang, Y. Zhao, Y. Xie, Y. Xu, L. Qian, Y. Yang and Z. Zhao, Matter, 2025, 8, 101913.
152. J. Meng, X. Liu, C. Niu, Q. Pang, J. Li, F. Liu, Z. Liu and L. Mai, Chem. Soc. Rev., 2020, 49, 3142–3186.
153. S. Fajal, D. Ghosh, K. Biswas, W. Mandal, N. Sarkar, G. K. Dam, A. Roy, A. Roychowdhury, D. Majumder and R. R. Urkude, Mater. Horiz., 2025, 12, 802–813.
154. Q. Zheng, A. Ren, A. Zagalskaya, H. Mao, D. Lee, C. Yang, K. C. Bustillo, L. F. Wan, T. A. Pham and J. A. Reimer, J. Am. Chem. Soc., 2024, 146, 34167–34175.
155. C. Wang, Y. Yao, J. Li and Y. Yamauchi, Acc. Mater. Res., 2022, 3, 426–438.
156. C. Wang, Q. Lai, K. Feng, P. Xu, X. Li and H. Zhang, Nano Energy, 2018, 44, 240–247.
157. X. Hui, X. Ge, R. Zhao, Z. Li and L. Yin, Adv. Funct. Mater., 2020, 30, 2005190.
158. J. Zou, J. Wu, Y. Wang, F. Deng, J. Jiang, Y. Zhang, S. Liu, N. Li, H. Zhang and J. Yu, Chem. Soc. Rev., 2022, 51, 2972–2990.
159. X. Li, N. Li, Z. Huang, Z. Chen, Y. Zhao, G. Liang, Q. Yang, M. Li, Q. Huang and B. Dong, ACS Nano, 2021, 15, 1718–1726.
160. J. Guo, G. Ma, G. Liu, C. Dai and Z. Lin, Adv. Energy Mater., 2024, 14, 2304516.
161. S. Li and A. S. Barnard, Chem. Mater., 2022, 34, 4964–4974.
162. W. Wu, S. Xu, Z. Lin, L. Lin, R. He and X. Sun, Energy Storage Mater., 2022, 49, 11–18.
163. Y. Zhou, Y. Fan, J. Zhang, C. Liao, M. Su, T.-S. Chan, Y.-R. Lu, Y.-C. Chuang, G.-Y. A. Tan and K. Shih, Sep. Purif. Technol., 2021, 278, 119595.
164. H. Zhu, W. Zhou and J. A. Yarmoff, J. Phys. Chem. C, 2018, 122, 16122–16131.
165. P. Stanley, J. Haimerl, N. Shustova, R. Fischer and J. Warnan, Nat. Chem., 2022, 14, 1342–1356.
166. B. Yuan, Q. Xu, X. Zhao, S. Geng, S. Tang, C. Zhang and H. Sun, Adv. Energy Mater., 2024, 14, 2303127.
167. J. Zhao, Y. Qi, Y. Chen, M. Zhang, Z. An, W. Wang, B. Nian, Y. Li, S. Han and L. Zhang, Angew. Chem., Int. Ed., 2025, 137, e202421905.
168. S. Neodo, D. Rosestolato, S. Ferro and A. D. Battisti, Electrochim. Acta, 2012, 80, 282–291.
169. J. Zhang, M. Shi, X. Ren, C. Wu, S. Hu, J. Feng, G. Li and C. Peng, J. Energy Storage, 2024, 88, 111604.
170. Z. Shao, J. Chen, Q. Xie and L. Mi, Coord. Chem. Rev., 2023, 486, 215118.
171. Y. Tulchinsky, C. H. Hendon, K. A. Lomachenko, E. Borfecchia, B. C. Melot, M. R. Hudson, J. D. Tarver, M. D. Korzynski, A. W. Stubbs and J. J. Kagan, J. Am. Chem. Soc., 2017, 139, 5992–5997.
172. Z. Lv, R. Lin, Y. Yang, K. Lan, C.-T. Hung, P. Zhang, J. Wang, W. Zhou, Z. Zhao and Z. Wang, Nat. Chem., 2025, 17, 177–185.
173. P. S. Mandal, G. Springholz, V. V. Volobuev, O. Caha, A. Varykhalov, E. Golias, G. Bauer, O. Rader and J. Sánchez-Barriga, Nat. Commun., 2017, 8, 968.
174. B. Wang, S. Zhong, P. Xu and H. Zhang, J. Mater. Chem. C, 2020, 8, 15526–15574.
175. C. H. Lin, L. Hu, X. Guan, J. Kim, C. Y. Huang, J. K. Huang, S. Singh and T. Wu, Adv. Mater., 2022, 34, 2108616.
176. S. Khatun, H. Bhunia and A. J. Pal, Phys. Chem. Chem. Phys., 2018, 20, 17934–17941.
177. H. Liu, C. Y. Chen, H. Yang, Y. Wang, L. Zou, Y. S. Wei, J. Jiang, J. Guo, W. Shi and Q. Xu, Adv. Mater., 2020, 32, 2004553.
178. K. Zhang, Y. Ge, Q. Yu, P. Zhang, Y. Feng, Z. Tie, J. Ma and Z. Jin, Energy Storage Mater., 2024, 67, 103296.
179. Y. Wang, X. Jin, J. Xiong, Q. Zhu, Q. Li, R. Wang, J. Li, Y. Fan, Y. Zhao and X. Sun, Adv. Mater., 2024, 36, 2404093.
180. Z. Zhang, Y. Li, F. Mo, J. Wang, W. Ling, M. Yu and Y. Huang, Adv. Mater., 2024, 36, 2311914.
181. S. Chen, S. Li, L. Ma, Y. Ying, Z. Wu, H. Huang and C. Zhi, Angew. Chem., Int. Ed., 2024, 63, e202319125.
182. C. Xu, C. Lei, P. Jiang, W. Yang, W. Ma, X. He and X. Liang, Joule, 2024, 8, 461–481.
183. S. Chen, C. Peng, D. Zhu and C. Zhi, Adv. Mater., 2024, 36, 2409810.
184. W. Li, H. Xu, S. Ke, H. Zhang, H. Chen, G. Guo, X. Xiong, S. Zhang, J. Fu and C. Jing, Nano-Micro Lett., 2025, 17, 143.
185. X. Zheng, Z. Liu, J. Sun, R. Luo, K. Xu, M. Si, J. Kang, Y. Yuan, S. Liu and T. Ahmad, Nat. Commun., 2023, 14, 76.
186. Y. Liu, L. Zhang, L. Liu, Q. Ma, R. Wang, P. Xiong, H. Li, S. Zhang, J. Hao and C. Zhang, Adv. Mater., 2025 DOI:10.1002/adma.202415979.
187. M. Wang, Y. Meng, M. Sajid, Z. Xie, P. Tong, Z. Ma, K. Zhang, D. Shen, R. Luo and L. Song, Angew. Chem., Int. Ed., 2024, 136, e202404784.
188. W. Li, H. Xu, H. Zhang, F. Wei, T. Zhang, Y. Wu, L. Huang, J. Fu, C. Jing and J. Cheng, Energy Environ. Sci., 2023, 16, 4502–4510.
189. H. Wu, J. Hao, S. Zhang, Y. Jiang, Y. Zhu, J. Liu, K. Davey and S.-Z. Qiao, J. Am. Chem. Soc., 2024, 146, 16601–16608.
190. S. Zeng, S. Chen, Z. Ao, X. Lin, L. Yan, C. Liu and Z. Lin, Small, 2025, 21, 2501695.
191. F. Yang, J. Long, J. A. Yuwono, H. Fei, Y. Fan, P. Li, J. Zou, J. Hao, S. Liu and G. Liang, Energy Environ. Sci., 2023, 16, 4630–4640.
192. X. Kong, J. Zhang, X. Zhang, Z. Wang and D. Wang, ACS Appl. Energy Mater., 2025, 8, 601–610.
193. S. Liu, L. Kang and S. C. Jun, Adv. Mater., 2021, 33, 2004689.
194. W. Lyu, X. Yu, Y. Lv, A. M. Rao, J. Zhou and B. Lu, Adv. Mater., 2024, 36, 2305795.
195. L. Hu, P. Xiao, L. Xue, H. Li and T. Zhai, EnergyChem, 2021, 3, 100052.
196. J. Yang, Q. Dai, S. Hou, C. Han and L. Zhao, Adv. Mater., 2025, 37, 2418258.
197. W. Tian, H. Zhang, X. Duan, H. Sun, G. Shao and S. Wang, Adv. Funct. Mater., 2020, 30, 1909265.
198. A. Stein, Z. Wang and M. A. Fierke, Adv. Mater., 2009, 21, 265–293.
199. L. Zhang and M. Jaroniec, Chem. Soc. Rev., 2020, 49, 6039–6055.
200. L. Zeng, J. Zhu, P. K. Chu, L. Huang, J. Wang, G. Zhou and X. F. Yu, Adv. Mater., 2022, 34, 2204636.
201. A. Vu, Y. Qian and A. Stein, Adv. Energy Mater., 2012, 2, 1056–1085.
202. C. Wang, X. Li, X. Xi, W. Zhou, Q. Lai and H. Zhang, Nano Energy, 2016, 21, 217–227.
203. Y. Wu, Y. Qian, C. Huang, Y. Zhang, Y. Yang, A. Hu, Q. Tang and X. Chen, Electrochim. Acta, 2023, 460, 142593.
204. Y. Xu, Y. Li, H. Jia, B. Liu, C. Wang and L. Li, Mater. Lett., 2023, 353, 135241.
205. N. S. Alghamdi, M. Rana, X. Peng, Y. Huang, J. Lee, J. Hou, I. R. Gentle, L. Wang and B. Luo, Nano-Micro Lett., 2023, 15, 209.
206. L. Tang, C. Yuan, W. Lu and X. Li, Adv. Funct. Mater., 2025 DOI:10.1002/adfm.202502455.
207. K. Lu, Z. Hu, J. Ma, H. Ma, L. Dai and J. Zhang, Nat. Commun., 2017, 8, 527.
208. W. Ye, H. Wang, J. Shen, S. Khan, Y. Zhong, J. Ning and Y. Hu, Chin. Chem. Lett., 2023, 34, 107198.
209. X. Jin, L. Song, C. Dai, Y. Xiao, Y. Han, X. Li, Y. Wang, J. Zhang, Y. Zhao and Z. Zhang, Adv. Mater., 2022, 34, 2109450.
210. J. Wang, H. Xu, R. Zhang, G. Sun, H. Dou and X. Zhang, Nanoscale, 2024, 16, 6596–6602.
211. W. Qu, C. Wen, B. Chen, Y. Cai and M. Zhang, Sci. China Mater., 2024, 67, 2889–2897.
212. T. Huang, S. Wang, M. Wang, H. Hu, J. Wang, X. Su, S. Xiao, J. Wu and Y. Gao, Green Energy Environ., 2025 DOI:10.1016/j.gee.2025.05.002.
213. Y. Zhang, Q. Li, W. Feng, S. Gao, H. Yue, Y. Su, H. Zhou, J. Huang, L. Han and M. Shakouri, Adv. Mater., 2025, 37, 2507609.
214. H. Wang, Y. Shao, S. Mei, Y. Lu, M. Zhang, J.-K. Sun, K. Matyjaszewski, M. Antonietti and J. Yuan, Chem. Rev., 2020, 120, 9363–9419.
215. X. Wang, G. Sun, P. Routh, D.-H. Kim, W. Huang and P. Chen, Chem. Soc. Rev., 2014, 43, 7067–7098.
216. T. Liu, H. Wang, C. Lei, Y. Mao, H. Wang, X. He and X. Liang, Energy Storage Mater., 2022, 53, 544–551.
217. J. Ye, W. Tian, Y. Du and J. Ji, Adv. Funct. Mater., 2025 DOI:10.1002/adfm.202509582.
218. T.-C. Chang, J.-P. Zhang and Y.-K. Fuh, J. Power Sources, 2014, 245, 66–75.
219. Q. Zhang, H. Jiang, S. Liu, Q. Wang, J. Wang, Z. Zhou, K. Cai, Q. Lai and Q. Wang, J. Power Sources, 2024, 601, 234286.
220. W. Cao, T. Hu, Y. Zhao, Z. Li, Y. Hu, C. Li, Y. Sun, F. Ding and G. Liang, ACS Energy Lett., 2024, 10, 320–329.
221. S. Zhang, J. Huang, L. Ma, D. Zhai, B. Wei, H. Yang and C. He, Energy Rev., 2024, 3, 100075.
222. W. Feng, Y. Wang, F. Tian, Z. Liu, X. Wei, C. Ma, G. Ma, Z. Li, D. Kong and L. Zhi, Energy Storage Mater., 2024, 73, 103812.
223. S. Qi, L. Dai, W. Huo, Y. Jiang, S. Yuan, Y. Xiao, Y. Liu, L. Wang and Z. He, Composites, Part B, 2023, 265, 110947.
224. L. Ma, G. Zhu, Z. Wang, A. Zhu, K. Wu, B. Peng, J. Xu, D. Wang and Z. Jin, Nano Lett., 2023, 23, 5272–5280.
225. W. Qu, J. Zhu, G. Cao, S. Chen, Y. Tan, B. Chen and M. Zhang, Small, 2024, 20, 2310475.
226. L. Zhu, X. Guan, Y. Fu, Z. Zhang, Y. Li, Q. Mai, C. Zhang, Z. Yuan, Y. Wang and P. Li, Adv. Funct. Mater., 2024, 34, 2409099.
227. P. Jia, Y. Yu, T. Chen and H. Huang, Angew. Chem., Int. Ed., 2025, 64, e202508809.
228. J. Sun, Z. Wang, J. Zhang and D. Wang, J. Energy Storage, 2024, 90, 111716.
229. Y. Li, L. Li, W. Xu, Y. Zhong and X. Pu, J. Power Sources, 2024, 599, 234208.
230. X. Yang, L. Xu and Y. Li, Coord. Chem. Rev., 2024, 516, 215961.
231. L. H. Pei, D. M. Xu, Y. Z. Luo, S. J. Guo, D. R. Liu, S. J. Jiang, W. J. Zhang and F. F. Cao, Adv. Mater., 2025, 37, 2420005.
232. L. Li, K. Yuan and Y. Chen, Acc. Mater. Res., 2022, 3, 584–596.
233. Y. Ying, X. Luo, J. Qiao and H. Huang, Adv. Funct. Mater., 2021, 31, 2007423.
234. C. Dong, Y. Yu, C. Ma, C. Zhou, J. Wang, J. Gu, J. Ji, S. Yang, Z. Liu and X. Xu, Energy Environ. Sci., 2025, 18, 3014–3025.
235. P. Hei, Y. Sai, W. Li, J. Meng, Y. Lin, X. Sun, J. Wang, Y. Song and X. X. Liu, Angew. Chem., Int. Ed., 2024, 63, e202410848.
236. H. Chun, J. R. Lunger, J. K. Kang, R. Gómez-Bombarelli and B. Han, npj Comput. Mater., 2024, 10, 246.
237. S. Liu, H. Zhang, X. Peng, J. Chen, L. Kang, X. Yin, Y. Yusuke and B. Ding, ACS Nano, 2025, 19, 13591–13636.
238. S. Liu, L. Kang, J. Hu, E. Jung, J. Zhang, S. C. Jun and Y. Yamauchi, ACS Energy Lett., 2021, 6, 3011–3019.
239. H. Yu, M. Dai, J. Zhang, W. Chen, Q. Jin, S. Wang and Z. He, Small, 2023, 19, 2205767.
240. L. Kang, S. Liu, Q. Zhang, J. Zou, J. Ai, D. Qiao, W. Zhong, Y. Liu, S. C. Jun and Y. Yamauchi, ACS Nano, 2024, 18, 2149–2161.
241. Y. Zhang, L. Wang, Q. Li, B. Hu, J. Kang, Y. Meng, Z. Zhao and H. Lu, Nano-Micro Lett., 2022, 14, 208.
242. J. Kang, C. Wang, Z. Liu, L. Wang, Y. Meng, Z. Zhai, J. Zhang and H. Lu, Energy Storage Mater., 2024, 68, 103367.
243. L. Li, R. Li, S. Zhou, W. Xu, Y. Li, J. Zhang, L. Gao and X. Pu, Carbon Neutralization, 2024, 3, 222–232.
244. Y. Huang, F. Rehman, M. Tamtaji, X. Li, Y. Huang, T. Zhang and Z. Luo, J. Mater. Chem. A, 2022, 10, 5813–5834.
245. C. He, R. Wang and Y. Li, ACS Appl. Mater. Interfaces, 2024, 16, 57040–57050.
246. T. Hu, Y. Zhao, Y. Yang, H. Lv, R. Zhong, F. Ding, F. Mo, H. Hu, C. Zhi and G. Liang, Adv. Mater., 2024, 36, 2312246.
247. X. Liu, H. Wang, J. Zhong, Z. Ma, W. Liu, R. Zhang, M. Li, F. Jiang and L. Kang, J. Energy Storage, 2024, 84, 110765.
248. L. Aldon, A. Perea, M. Womes, C. Ionica-Bousquet and J.-C. Jumas, J. Solid State Chem., 2010, 183, 218–222.
249. L. Zhou, A. M. Yao, Y. Wu, Z. Hu, Y. Huang and Z. Hong, Adv. Theory Simul., 2021, 4, 2100196.
250. W. Zong, J. Li, C. Zhang, Y. Dai, Y. Ouyang, L. Zhang, J. Li, W. Zhang, R. Chen and H. Dong, J. Am. Chem. Soc., 2024, 146, 21377–21388.
251. X. Chen, C. Liu, X. Bai, J. Zhang, X. Chang, L. Hou, H. Huang, Y. Wei, B. Wu and W. Liu, Energy Storage Mater., 2025, 75, 103984.
252. D. Spurling, H. Krüger, N. Kohlmann, F. Rasch, M. P. Kremer, L. Kienle, R. Adelung, V. Nicolosi and F. Schütt, Energy Storage Mater., 2024, 65, 103148.
253. U. Ryu, S. Jee, P. C. Rao, J. Shin, C. Ko, M. Yoon, K. S. Park and K. M. Choi, Coord. Chem. Rev., 2021, 426, 213544.
254. H. Lin, M. Ma, H. Qi, X. Wang, Z. Xing, A. Alowasheeir, H. Tang, S. C. Jun, Y. Yamauchi and S. Liu, Prog. Mater. Sci., 2025, 151, 101427.
255. M. A. Zaed, K. H. Tan, N. Abdullah, R. Saidur, A. K. Pandey and A. M. Saleque, Open Ceramics, 2024, 17, 100526.
256. H. Bi, D. Tian, Z. Zhao, Q. Yang, Y. Yuan, R. Zhang, L. Ai, X. Wang and J. Qiu, Adv. Funct. Mater., 2025, 35, 2423115.
257. A. Vest, Meeting the Challenge of Future Commercial Energy Storage with TETRA PureFlow Ultra-Pure Zinc Bromide, (https://onetetra.com/blog/power-storage-batteries-with-tetra-pureflow-ultra-pure-zinc-bromide/).
258. H. Lemontree, Energy Storage settled in Beipiao Economic Development Zone to create a new highland for zinc-bromine liquid flow energy storage battery industry, (https://en.eeworld.com.cn/news/newenergy/eic667255.html).

---