Sulfonated Cellulose Acetate Nanofibers Induced Zincophilic‐Hydrophobic Interface to Regulate Ion Transport for Long‐Lifespan Zinc‐Iodine Batteries

TL;DR

This paper introduces a protective layer made of sulfonated cellulose acetate nanofibers to improve the performance and lifespan of zinc-iodine batteries.

Contribution

The novel zincophilic-hydrophobic interface design using SCA nanofibers enhances ion transport and prevents battery degradation.

Findings

The SCA nanofiber membrane reduces water activity and promotes Zn2+ desolvation.

The membrane's electrostatic repulsion with polyiodide inhibits the shuttling effect.

Zinc-iodine batteries achieved over 13,000 cycles with high capacity retention and reversibility.

Abstract

Aqueous zinc‐iodine batteries (ZIBs) have attracted extensive attention due to their advantages of high theoretical specific capacity, abundant reserves, high safety, and low cost, while the Zn anodes are still suffering from dendrite growth, side reactions, and polyiodide corrosion, seriously affecting the service life of ZIBs. Herein, sulfonated cellulose acetate (SCA) nanofiber membrane with zincophilic‐hydrophobic property is constructed on the Zn anode as a protective layer by electrospinning to circumvent the above problems and achieve a stable Zn anode. Attributing to both the hydrophobicity and zincophilicity, the SCA nanofiber membrane not only reduces the activity of water but also promotes the Zn2+ desolvation. Moreover, negatively‐charged groups of the SCA nanofiber membrane cause electrostatic repulsion with polyiodide. Density functional theory calculations and COMSOL simulations further reveal that the SCA nanofiber membrane can tune the uniform 3D deposition behavior of Zn2+ by chemisorption and physical structure, respectively. The obtained ZIBs can achieve ultra‐long life span (> 13000 cycles) with high‐capacity retention (96.74%) and reversibility (average CE: 99.83%), demonstrating the reliability of our proposed strategy for achieving stable and high‐performance ZIBs.

Sulfonated cellulose acetate (SCA) nanofiber membrane with zincophilic‐hydrophobic property is constructed on the surface of the Zn anode by the electrospinning technique to tune the 3D deposition behavior of Zn2+ by chemisorption and micro‐sized physical structure. The negatively‐charged groups of SCA nanofiber membrane form an electrostatic repulsion force with polyiodide, which displays a certain inhibitory effect on the shuttling of polyiodide.

Figures

(a) Schematic illustration of the function mechanism of the SCA nanofiber membrane as a protective layer of the Zn anode. (b) XRD patterns of bare Zn and SCA@Zn electrodes. (c) The SEM image of the SCA nanofiber membrane and (d) corresponding EDS mapping images. High resolution XPS spectra of (e) C 1s, and (f) O 1s for SCA@Zn. (g) Contact angle measurements of the electrolyte on bare Zn and SCA@Zn.

(a) Cycling performance of bare Zn and SCA@Zn symmetric cells at 1 mA cm−2 with a capacity of 1 mAh cm−2. Insets are the detailed voltage profiles at 142 h (left) and 1000 h (right). (b) XRD patterns of bare Zn and SCA@Zn electrodes after 50 cycles. SEM images for the surface of (c) bare Zn anode, (d) SCA nanofiber membrane of SCA@Zn anode, and (e) Zn surface of SCA@Zn anode by removing the SCA nanofiber membrane after 50 cycles at 1 mA cm−2. In situ OM images of Zn deposits on (f) bare Zn and (g) SCA@Zn in symmetric Zn cells (The scale bars are 500 µm). AFM images of (h) bare Zn, and (i) SCA@Zn anodes after 50 cycles at 1 mA cm−2.

(a) SEM images of Zn foil after soaking in polyiodide solution for 3 days. (b) Cycling performance of Zn symmetric cells with mixed electrolyte and 2 m ZnSO4 electrolyte at a current density of 1 mA cm−2 with a capacity of 1 mAh cm−2. Optical images of H‐type electrolytic cells with polyiodide solution using (c) GF separator and (d) SCA nanofiber membrane. (e) Cycling performance of SCA@Zn symmetric cell with mixed electrolyte at a current density of 1 mA cm−2 with a capacity of 1 mAh cm−2. Insets are the detailed voltage profiles at 10 h (left) and 450 h (right). (f) Zeta potential values of CA and SCA.

Adsorption energy of Zn2+ on (a) bare Zn and (b) SCA with corresponding structural models (Insets with the gray, white, red, yellow, and blue balls representing carbon, hydrogen, oxygen, sulfur, and zinc atoms, respectively) for DFT calculations. (c) Chronoamperometry curve and the corresponding EIS plots (Inset) of the SCA@Zn symmetric cell. (d) Tafel plots of corrosion behaviors and (e) LSV curves for bare Zn and SCA@Zn. The growth of dendrites on (f) bare Zn and (g) SCA@Zn by COMSOL Multiphysics simulations. (h) Chronoamperometry curves of bare Zn and SCA@Zn at a −150 mV overpotential. The insets are the schematics of the Zn2+ diffusion and reduction processes on bare Zn and SCA@Zn.

(a) CV curves of Zn//ACI and SCA@Zn//ACI full cells at 0.1 mV s−1. (b) EIS plots of Zn//ACI and SCA@Zn//ACI full cells before cycling. (c) Galvanostatic charge/discharge curves at current densities from 0.2 to 4.2 A g−1. (d) Long‐term cycling stabilities of Zn//ACI and SCA@Zn//ACI full cells at 4.2 A g−1. SEM images for the surfaces of (e) bare Zn and (f) SCA@Zn (removing SCA nanofiber membrane) after 1000 cycles.

Full text

Introduction

1

In recent years, lithium‐based energy storage devices represented by lithium‐ion batteries (LIBs) have occupied most of the market [1, 2]. However, the high cost and flammability of lithium greatly hinder their practical applications [3]. Aqueous zinc‐based batteries, featuring high safety, resource availability, and high theoretical gravimetric capacity, have received extensive attention [4, 5, 6]. In contrast to zinc‐ion battery with ion intercalation/extraction [7, 8, 9], aqueous rechargeable zinc‐iodine batteries (ZIBs) can store/release energy based on the redox reaction of iodine at the cathode and the deposition/stripping of zinc at the anode. The multiple valence states of iodine endow it with unparalleled advantage for fascinating multi‐electron conversion reactions theoretically [10, 11, 12]. Although ZIBs are considered to be excellent candidates for next‐generation energy storage devices, they are also plagued by the tough issues to be addressed urgently. The major problems of I_2_ cathode lie on its poor electrical conductivity and the formation of soluble polyiodide [13, 14]. The as‐generated polyiodide species can diffuse and penetrate through the separator easily, which is known as the polyiodide shuttling effect. Besides, the hydrogen evolution reaction (HER) also occurs in a weakly acidic ZnSO_4_ solution. Accompanied by HER process, the pH will gradually increase, resulting in the formation of by‐products and an irregular interfacial electric field [15, 16]. Significantly, the above parasitic reactions terminate on the Zn anode, directly affecting its stability. On the one hand, Zn anodes are susceptible to the dendrite formation as a result of the uneven surface, leading to poor cycling performance and Coulombic efficiency (CE) [17, 18, 19]. On the other hand, polyiodide shuttles to the Zn anode, exacerbating metal corrosion and passivation, further damaging the electrode structure [20, 21, 22]. The above‐mentioned issues seriously limit the industrial application of ZIBs [23, 24]. Consequently, it is crucial to explore effective and implementable strategies for obtaining a highly stable and dendrite‐free Zn anode [25, 26], including the cathode design [27, 28, 29], the modulation of electrolyte composition [30, 31, 32, 33], the development of advanced separators [34, 35, 36], and the construction of protective layers on the Zn anode [37, 38, 39].

Among these strategies, constructing a protective layer on the Zn anode is one of the most effective and practical approaches, as it can effectively reduce the Zn anode from contact with the electrolyte physically, thereby inhibiting side reactions and corrosion by polyiodide. In previous works, considerable efforts were devoted to developing hydrophilic protective layers, while the enriched free water molecules still inevitably participate in side reactions [40, 41, 42]. In order to inhibit the activity of water molecules, it is effective to design hydrophobic layers for the reduction of the occurrence of side reactions [43]. For instance, Liu et al. proposed a hydrophobic protective strategy to achieve a stable Zn metal anode with enhanced electrochemical performance [44]. Nevertheless, inhomogeneous Zn deposition is inevitable since the microenvironment (e.g., the concentrations of Zn^2+^ within the electrolyte and the region near the surface of Zn anode) is not constant and uniform, leading to the formation of Zn dendrites [18]. It is remarkable that the formed Zn dendrites provide more reaction sites for side reactions and the corresponding byproducts impede the uniform Zn deposition, accelerating the formation of Zn dendrites in turn. Accordingly, it is particularly important to stabilize the Zn anode within ZIBs by achieving uniform Zn deposition. Organic nanofiber membranes, characterized by high specific surface area and porosity [45, 46, 47], exhibit low areal current density and fast diffusion rate of Zn^2+^, thereby achieving a small polarization to obtain homogeneous Zn deposition and suppress the Zn dendrite formation effectively. As compared to the irreversible fracture of inorganic protective layers with relative brittleness during the charging and discharging processes, the high flexibility and adjustability of the organic protective layer can cope with the stresses generated by the Zn anode and guarantee the interface stability. Therefore, polymer‐based nanofiber membranes with hydrophobicity possess an enormous potential as protective layers for Zn anodes in ZIBs.

In addition to addressing the issues of dendrites and side reactions on the Zn anode, extensive studies have been dedicated to regulating the shuttle effect of polyiodide by constructing functional protective layers [44]. Ion‐selective layer that simultaneously blocks polyiodides and conducts Zn^2+^ has shown considerable effectiveness. For example, Zhang et al. [39] used a sulfonic acid‐rich ion exchange layer to regulate the transport of polyiodide and Zn^2+^ at the zinc/electrolyte interface, which can effectively protect the zinc anode from corrosion and achieve uniform zinc deposition. Inspired by this strategy, a nanofiber membrane bearing similar ion‐selective functionalities may also enhance cycling stability and long‐term durability synergistically.

Here, we propose functional nanofiber membranes as the protective layers for ZIBs. Cellulose acetate (CA), as an organic biomass material, is widely available, inexpensive, and electrochemically stable [48]. Therefore, sulfonated cellulose acetate (SCA) was obtained as the raw material to in situ construct protective layers with zincophilic‐hydrophobic property on the surface of the Zn anode by electrospinning for addressing the multiple issues, including dendrite growth, side reaction, and polyiodide shuttling. Specifically, the SCA nanofiber membrane can inhibit side reactions by weakening H_2_O activity, and guide the uniform deposition of Zn^2+^ by rapid ion transport and zincophilic groups simultaneously. Moreover, the SCA nanofiber membrane has a certain electrostatic repulsion to polyiodide because of the ‐SO_3_ ^−^ group. Density functional theory calculations and COMSOL simulations reveal that the SCA nanofiber membrane can tune the 3D deposition behavior of Zn^2+^ by chemisorption and micro‐sized control over the physical structure. The resulting ZIBs achieve ultralong lifespan (> 13000 cycles), as well as high‐capacity retention (96.74%) and reversibility at 4.2 A g^−1^.

Results and Discussion

2

To address the issues of Zn dendrites, side reactions, and polyiodide corrosion on the Zn anode within ZIBs, the strategy of using SCA nanofiber membranes with zincophilic‐hydrophobic properties as protective layers of the Zn anode was proposed, as illustrated in Figure 1a. The SCA nanofiber membrane is expected to achieve a stable Zn anode through its favorable 3D pore structure and negatively charged groups. The building‐up route of SCA is shown in Figure S1, with a detailed synthetic procedure outlined in the Experimental Section. Specifically, CA was sulfonated by sodium p‐styrenesulfonate hydrate to introduce ‐SO_3_ ^−^ groups along the polymer backbone. The chemical structures of CA and SCA were characterized by Fourier transform infrared (FTIR) spectra (Figure S2). As compared with the CA, the FTIR spectrum of SCA shows new peaks at 1125 and 1038 cm^−1^, which correspond to the asymmetric and symmetric stretching vibrations of O═S═O, respectively. To be noted, several drying/washing cycles were employed to remove any residual homopolymer formed during sulfonation. As depicted in Figure S3, a rough surface with minor scratches can be observed on the commercial Zn foil. An uneven electric field may be formed in this kind of surface during charging and discharging processes, resulting in the formation of severe Zn dendrites. To achieve a relatively flat surface, SCA nanofiber membranes were in situ constructed on the surface of the Zn anode by electrospinning, which possesses an adjustable structure and simple operation without additives compared to traditional coating methods. The X‐ray diffraction (XRD) patterns of the bare Zn foil and the Zn with SCA nanofiber membrane (SCA@Zn) are displayed in Figure 1b, in which the SCA@Zn shows the same characteristic diffraction peaks as those in Zn foil, suggesting the tight connection between the nanofiber membrane and Zn foil [49]. The optical images of the bare Zn and SCA@Zn are shown in Figure S4a,b, from which the thickness of the SCA nanofiber membrane is measured to be 100 µm (Figure S4c,d). Additionally, the microstructure morphology of the SCA nanofiber membrane recorded by scanning electron microscope (SEM) is shown in Figure 1c, which almost remains the same as the CA (Figure S5). The diameter of SCA nanofiber ranges from 0.2 to 1.0 µm, suggesting a relatively uniform and porous surface with nanoscale roughness. The energy‐dispersive X‐ray spectroscopy (EDS) mapping images of O and S elements (Figure 1d) also indicate the uniform distribution of the ─SO_3_ ^−^ group on the SCA nanofiber membrane. Additionally, X‐ray photoelectron spectroscopy (XPS) was carried out to characterize SCA@Zn as depicted in Figure 1e,f. In the C 1s XPS spectrum of SCA@Zn, three peaks at 284.9, 285.9, and 287 eV correspond to the C─C/C─H, C─O/C─OH, and O─C─O/O═C─O species, respectively. The O 1s spectrum of SCA@Zn can be divided into three characteristic peaks at 531.4, 532.5, and 534.1 eV, corresponding to the S═O, C─OH, and O═C─O species, respectively. The abundant polar functional groups in the SCA molecule can serve as abundant coordination sites to promote the desolvation of hydrated Zn^2+^ [50].

(a) Schematic illustration of the function mechanism of the SCA nanofiber membrane as a protective layer of the Zn anode. (b) XRD patterns of bare Zn and SCA@Zn electrodes. (c) The SEM image of the SCA nanofiber membrane and (d) corresponding EDS mapping images. High resolution XPS spectra of (e) C 1s, and (f) O 1s for SCA@Zn. (g) Contact angle measurements of the electrolyte on bare Zn and SCA@Zn.

Apart from the uniform surface, the SCA nanofiber membrane can functionalize the surface of Zn foil with enhanced hydrophobicity due to the unique pore structure, which is formed by the overlapped nanofibers observed in an atomic force microscope (AFM) image (Figure S6). These pore sizes of the SCA nanofiber membrane are much smaller than the diameter of a minimum water droplet visible to the naked eye (100 µm), while the pores of the SCA nanofiber membrane contain a large amount of air. When the water droplet contacts the fiber membrane, therefore, a water–air interface will be formed, as the Cassie model displayed in Figure S7 [51]. This air cushion reduces the solid–liquid contact area and increases the apparent contact angle. To prove this conjecture, the SCA film without a nanofibrous structure obtained by direct coating on Zn anode has a relatively small contact angle of 77° (Figure S8), in stark contrast to the 111° contact angle of SCA@Zn (Figure 1g), demonstrating the hydrophobicity of the SCA@Zn surface caused by its unique nanoporous structure. Notably, the contact angle of the electrolyte on the SCA@Zn surface was measured at different contact times. According to the results (Figure S9), the contact angle hardly changes within 20 min, proving the extraordinary hydrophobicity of the SCA@Zn surface. Additionally, the contact angle on the surface of bare Zn foil is 98° (Figure 1g), which shows a decrease relative to that of SCA@Zn, demonstrating the positive role of SCA nanofiber membrane in increasing the hydrophobicity of the surface of Zn anode and reducing water molecules from reaching the zinc surface [43].

The surface of bare Zn can be easily corroded in the ZnSO_4_ electrolytes because of the side reactions [52]. To further evaluate the corrosion resistance of the SCA nanofiber membrane, bare Zn and SCA@Zn were immersed in 2 m ZnSO_4_ electrolytes for several days. Comparing the XRD patterns of bare Zn foil and SCA@Zn (Figure S10), new diffraction peaks at 16.1° and 24.4° can be observed on the bare Zn after soaking for 1 day, which correspond to the planes from the byproducts of Zn_4_SO_4_(OH)6·5H_2_O and indicate the initial corrosion of the Zn surface. On the contrary, no characteristic peaks from related byproducts of Zn_4_SO_4_(OH)6·5H_2_O can be detected on the SCA@Zn after immersed for even 7 days, indicating its excellent chemical stability and corrosion resistance. The symmetric cells were first fabricated to evaluate the plating/stripping cycling stability of the Zn electrode protected by the SCA nanofiber membrane. The voltage curve of the symmetric cell at a current density of 1 mA cm^−2^ (1 mAh cm^−2^) is shown in Figure 2a. A sharp voltage drop at 150 h is exhibited on the symmetric cell with a bare Zn electrode, which is related to the accumulation of byproducts and severe dendrite growth. However, the symmetric battery with SCA@Zn electrode exhibits a long‐term cycling life of more than 1050 h, which is approximately 7 times that of the bare Zn cell. The bare Zn and SCA@Zn electrodes after 50 cycles were characterized by the XRD technique. In contrast to the bare Zn with an obvious difference, almost no additional diffraction peak can be observed on the cycled SCA@Zn electrode (Figure 2b), demonstrating that the SCA nanofiber membrane endows excellent corrosion resistance. Even at a higher areal capacity of 2 mAh cm^−2^, the cell with SCA@Zn electrode remains stable within about 290 h (Figure S11). The electrochemical impedance spectrum (EIS) plots of bare Zn and SCA@Zn symmetric cells are provided in Figure S12, displaying that the resistance of the SCA@Zn is slightly higher than that of the bare Zn. Furthermore, the SCA@Zn//Cu half cell can cycle stably for more than 200 cycles with an average CE of 98.32% at 1 mA cm^−2^ for 1 mAh cm^−2^ (Figure S13). As the voltage‐capacity profiles shown in Figure S14, the SCA@Zn//Cu still displays inconspicuous fluctuation after 200 cycles, indicating the enhanced reversibility of the Zn stripping/deposition process by constructing the SCA nanofiber membrane on the surface of the Zn anode.

(a) Cycling performance of bare Zn and SCA@Zn symmetric cells at 1 mA cm−2 with a capacity of 1 mAh cm−2. Insets are the detailed voltage profiles at 142 h (left) and 1000 h (right). (b) XRD patterns of bare Zn and SCA@Zn electrodes after 50 cycles. SEM images for the surface of (c) bare Zn anode, (d) SCA nanofiber membrane of SCA@Zn anode, and (e) Zn surface of SCA@Zn anode by removing the SCA nanofiber membrane after 50 cycles at 1 mA cm−2. In situ OM images of Zn deposits on (f) bare Zn and (g) SCA@Zn in symmetric Zn cells (The scale bars are 500 µm). AFM images of (h) bare Zn, and (i) SCA@Zn anodes after 50 cycles at 1 mA cm−2.

To deeply unveil the relationship between structure and property, SEM images were recorded and displayed in Figure 2c–e. After 50 cycles, obvious flaky dendrites are observed on the bare Zn (Figure 2c), exacerbating the short circuit of the cell. In sharp contrast, the SCA nanofiber membrane maintains the original morphology without any significant dendrites and byproducts (Figure 2d). The main reason is that the highly connected 3D ion channels of the SCA nanofiber membrane make the diffusion of Zn^2+^ ions uniform, avoiding the blockage of pores caused by local deposition. Simultaneously, a relatively dense and smooth surface morphology of Zn (after removing SCA nanofiber membrane) is exhibited in Figure 2e with corresponding low‐magnification SEM images shown in Figure S15, strongly demonstrating the capability of the SCA nanofiber membrane in guiding uniform Zn deposition on the surface of Zn anode and suppressing side reactions. To further observe the deposition of Zn enabled by the interface at a macroscopic scale, the morphology evolution of the Zn anode was studied by in situ optical microscopy (OM). As the OM images shown in Figure 2f, a small Zn dendrite appears on the surface of bare Zn after plating for 10 min, steadily growing and eventually developing into large dendrites. Simultaneously, the bubble is generated at 45 min, which corresponds to the occurrence of HER. Conversely, the SCA@Zn preserves a homogeneous and dense surface, without any bubbles and dendrites during the entire 45 min plating process (Figure 2g), demonstrating the ability of the SCA nanofiber membrane to inhibit the hydrogen evolution reaction and guide the uniform deposition of Zn^2+^ [53]. Additionally, the Zn surface was scanned by AFM after the cycling process, as depicted in Figure 2h,i. The dramatic height difference on the surface of bare Zn can probably be caused by severe corrosion and uneven Zn deposition [54], while the surface of Zn anode protected by a SCA nanofiber membrane is flat and illustrates the optimized Zn deposition behavior with SCA nanofiber membrane. The above results validate the superiority of the SCA nanofiber membrane in terms of suppressing the side reactions and dictating Zn deposition behavior.

Apart from the growth of Zn dendrites, the Zn anode can also be corroded by polyiodide species. To observe the corrosion degree of Zn anode from polyiodide species, Zn foils were put into ZnSO_4_ and polyiodide solutions (composed of 1 m KI and 0.01 m I_2_ in deionized water), respectively. As displayed in Figure 3a, a large honeycomb surface can be observed on the Zn foil in polyiodide solution, which is more corrosive than the Zn foil in ZnSO_4_ solution (Figure S16). This stark contrast confirms the accelerated corrosive action of polyiodides on Zn foil. As shown in Figure 3b, the bare Zn symmetric cell was tested in mixed electrolyte (composed of 90 vol% 2 m ZnSO_4_ and 10 vol% aforementioned polyiodide solution) and 2 m ZnSO_4_ electrolyte, respectively. The Zn symmetric cell with mixed electrolyte occurs a short circuit in a very short time at a current density of 1 mA cm^−2^, while the cell using ZnSO_4_ electrolyte can work over 150 h, confirming the exacerbated corrosion of Zn surface by polyiodide ions. To examine the charge repulsion of the SCA nanofiber membrane, the permeation of polyiodide species was evaluated in an H‐type electrolytic cell. The SCA nanofiber membrane and the commercial glass fiber (GF) separator were used as the separators of the H‐type cells, respectively, to observe the polyiodide diffusion process within two devices. The left side of the glass device with GF separator obviously turns yellow from transparent after 6 h, indicating the severe polyiodide shuttle (Figure 3c). By stark contrast, the electrolyte color maintains unchanged on the left side of the glass device with the SCA nanofiber membrane even after 24 h, indicating that the SCA nanofiber membrane can inhibit the polyiodide shuttle effectively (Figure 3d). To exclude the influence of hydrophobicity on the polyiodide shuttle, the CA nanofiber membrane was used as the separator of the H‐type cells (Figure S17). The device with the CA nanofiber membrane exhibits a more intense yellow color than the one with the SCA nanofiber membrane at 24 h, demonstrating the SCA nanofiber membrane has a more significant inhibitory effect on the shuttling of polyiodide. Electrochemical measurements reinforce this behavior. The symmetric cell with SCA@Zn electrode in the mixed electrolyte can be cycled steadily for more than 460 h (Figure 3e), which is much longer than the bare Zn (27 h, Figure 3b), demonstrating a good cycling stability performance by coating the SCA nanofiber membrane on the surface of the Zn anode. Zeta potential measurement was performed to reflect the surface charge of the SCA nanofiber membrane in Figure 3f. As compared with the Zeta potential of CA (−7.3 mV), the Zeta potential of SCA is more negative, as −26.7 mV, due to the introduction of ‐SO_3_ ^−^ groups. The electronegativity of the SCA nanofiber membrane can form electrostatic repulsion with polyiodide, which verifies that SCA nanofiber membrane has a certain inhibitory effect on the shuttling of polyiodide.

(a) SEM images of Zn foil after soaking in polyiodide solution for 3 days. (b) Cycling performance of Zn symmetric cells with mixed electrolyte and 2 m ZnSO4 electrolyte at a current density of 1 mA cm−2 with a capacity of 1 mAh cm−2. Optical images of H‐type electrolytic cells with polyiodide solution using (c) GF separator and (d) SCA nanofiber membrane. (e) Cycling performance of SCA@Zn symmetric cell with mixed electrolyte at a current density of 1 mA cm−2 with a capacity of 1 mAh cm−2. Insets are the detailed voltage profiles at 10 h (left) and 450 h (right). (f) Zeta potential values of CA and SCA.

Density functional theory (DFT) calculations were carried out to determine the adsorption energy of Zn^2+^ on various surfaces (Figure 4a,b). The simulation model describing the interaction between Zn^2+^ and SCA molecule is shown in Figure 4b, where Zn^2+^ is shown to preferably adsorb onto the ‐SO_3_ ^−^ group. As compared with the adsorption energy of Zn^2+^ on the bare Zn (−0.50 eV), the SCA molecule is more inclined to adsorb Zn^2+^ with a relatively more negative adsorption energy of −1.69 eV, confirming the enhanced zinc affinity of the SCA molecule. Furthermore, chronoamperometry and EIS tests were utilized to determine the transference numbers of Zn^2+^ (tZn2+) of the symmetric cell with SCA@Zn electrode. As displayed in Figure 4c, the tZn2+ of the SCA nanofiber membrane is calculated to be 0.75 based on the following equation [55], which is significantly higher than the commercial anion exchange membrane (tZn2+: 0.19) [56].

$$[eqn]$$

where ΔV is the voltage polarization applied, and I_0_ (I_S_) and R_0_ (R_S_) are the initial (steady) state current and resistance, respectively. It also proves that the SCA nanofiber membrane can accelerate the transport of Zn^2+^. The Tafel curve was also recorded to study the effect of the SCA nanofiber membrane on the corrosion resistance of the Zn anode. As shown in Figure 4d, the corrosion potential of the Zn foil with the protection of the SCA nanofiber membrane shifts positively, representing a reduced tendency for corrosion reaction. The result indicates the enhanced corrosion resistance of SCA@Zn, further proving that the corrosion of the Zn anode can be inhibited by the hydrophobic SCA nanofiber membrane [57]. Furthermore, linear sweep voltammetry (LSV) tests were performed to fully evaluate the HER activities on SCA@Zn and bare Zn electrodes. As shown in Figure 4e, the SCA@Zn shows a significantly reduced overpotential compared with the bare Zn anode at the same current density [58], demonstrating the efficiently inhibited the HER process by using the hydrophobic SCA nanofiber membrane, that is consistent with the in situ OM images results shown in Figure 2f,g.

Adsorption energy of Zn2+ on (a) bare Zn and (b) SCA with corresponding structural models (Insets with the gray, white, red, yellow, and blue balls representing carbon, hydrogen, oxygen, sulfur, and zinc atoms, respectively) for DFT calculations. (c) Chronoamperometry curve and the corresponding EIS plots (Inset) of the SCA@Zn symmetric cell. (d) Tafel plots of corrosion behaviors and (e) LSV curves for bare Zn and SCA@Zn. The growth of dendrites on (f) bare Zn and (g) SCA@Zn by COMSOL Multiphysics simulations. (h) Chronoamperometry curves of bare Zn and SCA@Zn at a −150 mV overpotential. The insets are the schematics of the Zn2+ diffusion and reduction processes on bare Zn and SCA@Zn.

In order to visually observe the effect of the SCA nanofiber membrane on the transport and deposition of Zn^2+^, COMSOL Multiphysics simulations were carried out for the simulation of dendrite growth on the zinc foil (Figure 4f,g). As the surface of commercial zinc foil inevitably exhibits defects (e.g., scratches), it is easy to form tips or dendrites in the electric field. Therefore, three protrusions are given first on the Zn foil to represent its uneven surface (Figure S18). In addition, the nanofiber membrane is known to have micrometer‐sized pores formed by overlapping fibers, and its longitudinal section shape is round or oval. Therefore, the distributions of fibers in the SCA membrane are simulated by circles and ellipses with different sizes and angles, where the sparse circles and ellipses are set to simulate the high porosity of the SCA nanofiber membrane. For the bare Zn, there is an inhomogeneous strong electric field around the protrusions on the Zn surface, leading to the charge accumulation and aggravated formation of Zn dendrites eventually (Figure 4f). On the contrary, no dendrites were formed at protrusions (Figure 4g), as the SCA nanofiber membrane with a 3D porous structure can accelerate the transport of Zn^2+^ and reduce the current density on the zinc foil, making the electric field distribution on the Zn anode more uniform. Therefore, the COMSOL simulations indicate that the SCA nanofiber membrane can inhibit the 2D diffusion of Zn^2+^ and tune the 3D deposition behavior of Zn^2+^, which is beneficial to avoiding the growth of Zn dendrites [59]. Moreover, the dense structure formed by the traditional coating method is also simulated by increasing the number of circles and ellipses (Figure S19a). The Zn anode with the dense structure presents uniform electric field distributions at the protrusions (Figure S19b). In contrast to the structure with high porosity, however, the Zn^2+^ concentration at the protective film is higher, which may be due to the slow transmission of Zn^2+^ in this region caused by the limited pores [60], indicating that the porous structure of the nanofiber membrane can adjust the electric field distribution more effectively than the dense structure. Additionally, chronoamperometry tests were performed at a constant overpotential of −150 mV to reflect the Zn deposition behavior, as depicted in Figure 4h. For the bare Zn anode, the current density keeps increasing to 500 s and indicates a 2D diffusion process of Zn^2+^ ions, during which the Zn^2+^ ions tend to deposit on the pre‐existing deposition sites, resulting in the gradual roughness of the morphology and the final formation of Zn dendrites. In a shark contrast, the SCA@Zn electrode transits to a 3D diffusion process, indicating that the diffusion and nucleation of Zn^2+^ can be well regulated by the SCA nanofiber membrane [61]. The chronoamperometry results are consistent with the COMSOL simulations, suggesting the importance of the SCA nanofiber membrane to guide 3D diffusion of Zn^2+^.

To verify the effectiveness of the SCA@Zn anode, the full cell was assembled with active carbon/I_2_ (ACI) as the cathode and 2 m ZnSO_4_ as the electrolyte. The cyclic voltammetry (CV) curves shown in Figure 5a reveal that both of bare Zn//ACI and SCA@Zn//ACI cells show two typical redox peaks corresponding to the reversible redox reaction of iodine species at a scan rate of 0.1 mV s^−1^. Among them, the polarization potential of SCA@Zn (58 mV) is slightly higher than that of the bare Zn (32 mV) in the full cell, indicating that the SCA nanofiber membrane generates a limited effect on the redox process. As the EIS plots shown in Figure 5b, the charge transfer resistance of the cell with SCA@Zn is slightly larger than that of the cell with bare Zn, which is related to the overpotential induced by the SCA nanofiber membrane [39]. In Figure 5c, the charge/discharge curves of the cell with SCA@Zn anode are shown in the voltage range of 0.6–1.6 V with current densities from 0.2 to 4.2 A g^−1^. As the rate performance of the full cells presented in Figure S20, the average discharge capacity of the SCA@Zn//ACI cell is 147.6 mAh g^−1^ at 0.2 A g^−1^. When the current density increases to 4.2 A g^−1^, the SCA@Zn//ACI cell even maintains a reversible capacity of 101.2 mAh g^−1^ with a corresponding capacity retention of 68.6%. The long‐term cycling performance of the cells was also evaluated to investigate their potential for practical applications. Although the initial specific capacity of bare Zn//ACI is slightly higher, it decreases rapidly in a relatively short period of ≈5700 cycles (Figure 5d) due to the dendrite growth and polyiodide shuttling within its cell. Comparatively, the cycling performance of SCA@Zn//ACI is stable without obvious decay until 13000 cycles corresponding to a capacity retention of 96.74% and an average CE up to 99.83%. To unveil the advantages from the SCA nanofiber membrane, SEM images for the surface of bare Zn and SCA@Zn (the SCA nanofiber membrane was torn off) after 1000 cycles were observed. As shown in Figure 5e,f, the bare Zn anode displays a rough surface, while the Zn surface with the protection of SCA nanofiber membrane remains dense and flat. It indicates that the SCA nanofiber membrane can inhibit the shuttle of polyiodide to prevent the Zn anode from being corroded and prolong the service life of the cell [62].

(a) CV curves of Zn//ACI and SCA@Zn//ACI full cells at 0.1 mV s−1. (b) EIS plots of Zn//ACI and SCA@Zn//ACI full cells before cycling. (c) Galvanostatic charge/discharge curves at current densities from 0.2 to 4.2 A g−1. (d) Long‐term cycling stabilities of Zn//ACI and SCA@Zn//ACI full cells at 4.2 A g−1. SEM images for the surfaces of (e) bare Zn and (f) SCA@Zn (removing SCA nanofiber membrane) after 1000 cycles.

Furthermore, we also evaluated the stability of the SCA@Zn electrode under iodine‐rich conditions because the iodine species in the electrolyte can further accelerate Zn corrosion. A long cycle performance test was carried out at a current density of 2.1 A g^−1^ using activated carbon (AC) as the cathode and SCA@Zn as the anode, with 0.2 m KI in the electrolyte as iodine source (Figure S21). The increase in initial capacity would be derived from the gradual activation process of cells [63]. The cell with SCA@Zn anode is still very stable after 5000 cycles with a specific capacity of about 71 mAh g^−1^ (≈93.43% capacity retention). In contrast, the capacity of the battery without the SCA nanofiber membrane starts to decay rapidly after a short charge/discharge cycle, which is lower than that of SCA at about 3000 cycles. In terms of the above results, the presence of the SCA nanofiber membrane can greatly improve the stability of the energy storage reaction. Some reported protective layers are compared to our work, with the relevant details shown in Table S1. Our work exhibits competitive cycling stability both on current density and cycle life, demonstrating SCA nanofiber membrane is suitable for application in ZIBs. Simultaneously, the SCA nanofiber membrane has a large potential for application in ZIBs due to the advantages of low cost, simplicity, and convenience.

Conclusion

3

In summary, SCA nanofiber membranes with zincophilic‐hydrophobic property were constructed as protective layer on the surface of Zn anode by electrospinning. The hydrophobicity of the SCA nanofiber membrane reduces the activity of water molecules, inhibiting the occurrence of HER side reactions and byproducts. Theoretical calculations and experimental characterizations demonstrate that the uniform zincophilic ‐SO_3_ ^−^ groups exhibit strong adsorption with Zn^2+^ ions, promoting the Zn^2+^ desolvation and accelerating the transfer of Zn^2+^. Simultaneously, COMSOL simulations combined with related measurements reveal that the 3D nanoporous structure of the SCA nanofiber membrane provides abundant ion channels and induces the 3D diffusion of Zn^2+^, resulting in uniform and dense Zn deposition. Attributed to the electrostatic repulsion between ─SO_3_ ^−^ groups of SCA nanofiber membrane and polyiodide, the polyiodide shuttling is significantly confined, and thereby protecting the Zn anode from polyiodide ions. Consequently, ZIBs can achieve an ultralong lifespan (> 13000 cycles), as well as high‐capacity retention (96.74%), and coulombic efficiency (average CE: 99.83%) at a high current density of 4.2 A g^−1^. Therefore, this work not only provides a facile and easy‐to‐operate method for preparing low‐cost and multifunctional protective layers of Zn anode, but also offers a viable solution for achieving long‐life and high‐performance aqueous ZIBs.

Experimental Section

4

Materials

4.1

Cellulose acetate (CA) (acetyl: 39.8 wt.%) was purchased from Meryer (Shanghai) Biochemical Technology Co., Ltd. Potassium persulphate (AR), acetone (CP), ethanol (≥99.8%), and N,N‐dimethylacetamide (DMAc) were provided from Sinopharm Chemical Reagent Co., Ltd. Sodium p‐styrenesulfonate hydrate was provided by Tokyo Chemical Industry. ZnSO_4_ (99%) was obtained from Shanghai Aladdin Biochemical Technology Co., Ltd.

Preparation of Sulfonated Cellulose Acetate (SCA)

4.2

The dried CA (1 g) was dispersed in acetone (9 g) and stirred magnetically at 60°C for 3 h to obtain a transparent solution with a mass fraction of 10 wt.%. Subsequently, 0.277 mm potassium persulfate was added and stirred continuously to facilitate the free radical formation on CA backbone. After stirring for 10 min, 0.15 g of sodium p‐styrenesulfonate hydrate was added to the solution and reacted at 60°C for 3 h. After completing the grafting reaction, the grafted copolymers were precipitated using ethyl alcohol and then separated by filtration. Several drying/washing cycles were employed to get rid of homo‐polymer. Finally, the resulting grafted copolymer was dried in a vacuum oven at 60°C.

Preparation of SCA@Zn Electrode

4.3

Firstly, 9.4 mL of acetone was added into 5.6 mL of DMAc. Then, 1.4 g of SCA was added into the mixed solution and heated to 80°C for 6 h to obtain a spinning solution with a concentration of 10 wt.%. Electrospinning was performed at a voltage of 16 kV, with Zn foil wrapped around a roller for the collection of nanofibers. The distance between the tip of the needle and the collector was kept at 20 cm, and the polymer solution was injected at a controlled rate of 0.04 mL min^−1^. The electrospinning process was performed at 28°C and 45% humidity. Following the electrospinning process, the resulting SCA@Zn was dried at 60°C for 4 h to eliminate any residual solvent, thereby yielding the SCA@Zn electrode.

Preparation of Cathodes

4.4

Active carbon (AC, YP‐80F) and I_2_ were mixed and put at 80°C for 3 h to obtain a mixture of active carbon/I_2_ (ACI). The slurry was prepared by mixing ACI materials, carbon black, and polyvinylidene difluoride in a mass ratio of 8:1:1, where N‐methyl‐2‐pyrrolidone (NMP) was used as solvent. The slurry was coated on graphite paper and dried under a vacuum at 50°C overnight to fabricate the ACI cathode eventually. The mass loading of I_2_ is about 1.0 mg cm^−2^. To be noted, the AC cathode was fabricated in the same steps.

Materials Characterizations

4.5

Fourier transform infrared (FTIR) spectra were obtained on a Nicolet 6700 to check the chemical structures of materials. The crystal structures of materials were characterized using X‐ray diffraction (XRD) on a Bruker D8 ADVANCE spectrometer. X‐ray photoelectron spectroscopy (XPS, Escalab 250Xi) was conducted to analyze the chemical structures of the materials. Scanning electron microscopy (SEM) images and energy‐dispersive X‐ray spectroscopy (EDS) mappings were obtained on JSM‐7500F. Zeta potential analyses of materials were performed on SURPASS 3. The contact angle was evaluated by an optical surface analyser (OSA200).

Electrochemical Measurements

4.6

The CR2025‐type coin cells were assembled in the air to evaluate electrochemical performance. In the symmetric cells, identical Zn foils or SCA@Zn served as both of working and counter electrodes. For asymmetric cells, Cu foils were coupled with Zn or SCA@Zn. The full cells were assembled through employing ACI as the cathode, bare Zn foil or SCA@Zn as the anode. The applied current density was calculated based on the mass of I_2_. All galvanostatic charge/discharge measurements of cells were conducted using LAND instruments (CT2001A). For the three‐electrode system, bare Zn or SCA@Zn was used as the working electrode, graphite rod as the counter‐electrode, and saturated calomel electrode (SCE) as the reference electrode. Cyclic voltammetry (CV), electrochemical impedance spectrum (EIS), and chronoamperometry tests were carried out on DH7000C electrochemical workstation. Tafel and liner scan voltammetry (LSV) tests were carried out on CHI 760E electrochemistry workstation.

Density Functional Theory Calculations

4.7

All density functional theory (DFT) calculations were performed using the Vienna ab initio simulation package (VASP) based on plane‐wave basis sets and the projector augmented‐wave method. The exchange‐correlation potential was treated with the generalized gradient approximation (GGA) in the Perdew–Burke–Ernzerhof (PBE) parametrization. The van der Waals correction of Grimme's DFT‐D3 model was also adopted. A vacuum region of approximately 15 Å was applied to eliminate interactions between periodic images. The energy cutoff was set to be 450 eV. Brillouin‐zone integrations were sampled using a Γ‐centered Monkhorst–Pack k‐point mesh of 1 × 1 × 1. All structures were fully relaxed until the maximum force on each atom was less than 0.02 eV Å^−1^ and the energy convergence criterion reached 10^−5^ eV. The adsorption energy *E_ads_

• can be obtained by using the following equation:

$$[eqn]$$

where E_M_ stands for the energy of the monolayer with the adsorbed M molecule, E__ is the energy of the surface, and E_M_ is the energy of a M molecule under vacuum.

COMSOL Multiphysics Simulations

4.8

The 2D Zn symmetric cell was modeled based on COMSOL Multiphysics software as shown in Figures S18,19a, for both the simulations of electric field distribution and Zn^2+^ ionicity distribution. The protrusions on the Zn surface were represented by three semi‐ellipses, each with a long axis of 90 µm and a short axis of 50 µm. Circles and ellipses of different sizes and shapes were used to simulate fiber morphology. The diffusion coefficient of Zn^2+^ in the electrolyte domain was 1 ×  10^−9^ m^2^ s^−1^, while the diffusion coefficient of Zn^2+^ in two protective layers with different compactness was 1 ×  10^−7^ m^2^ s^−1^. “Tertiary current distribution” interface was selected to characterize the current and potential distributions in the cell and analyze the transport of Zn^2+^ species in the electrolyte, while the Butler–Volmer expression was used to reflect the electrode kinetics of the charge transfer reaction in the electrolyte at room temperature. Additionally, transformation geometry was further utilized to track the change of the Zn anode boundary with time during the electroplating process.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: advs73474‐sup‐0001‐SuppMat.docx.