Dendrite‐free Zn deposition initiated by nanoscale inorganic–organic coating‐modified 3D host for stable Zn‐ion battery

A 3D nanostructured scaffold as the host for zinc enables effective inhibition of anodic dendrite growth. However, the increased electrode/electrolyte interface area provided by using 3D matrices exacerbates the passivation and localized corrosion of the Zn anode, ultimately bringing about the degradation of the electrochemical performance. Herein, a nanoscale coating of inorganic–organic hybrid (α‐In2Se3‐Nafion) onto a flexible carbon nanotubes (CNTs) framework (ISNF@CNTs) is designed as a Zn plating/stripping scaffold to ensure uniform Zn nucleation, thus achieving a dendrite‐free and durable Zn anode. The introduced inorganic–organic interfacial layer is dense and sturdy, which hinders the direct exposure of deposited Zn to electrolytes and mitigates the side reactions. Meanwhile, the zincophilic nature of ISNF can largely reduce the nucleation energy barrier and promote the ion‐diffusion transportation. Consequently, the ISNF@CNTs@Zn electrode exhibits a low‐voltage hysteresis and a superior cycling life (over 1500 h), with dendrite‐free Zn‐plating behaviors in a typical symmetrical cell test. Additionally, the superior feature of ISNF@CNTs@Zn anode is further demonstrated by Zn‐MnO2 cells in both coin and flexible quasi‐solid‐state configurations. This work puts forward an inspired remedy for advanced Zn‐ion batteries.

1][32] However, the utilization of a 3D host for Zn deposition still faces some intractable challenges.First, the porous skeleton of these 3D hosts benefits from adequate contact between the electrode and electrolyte interfaces, which will significantly increase the contract area that can accelerate the Zn corrosion. 9Second, Zn is found to be deposited on the top rather than inside parts of the host, especially at a high current densities, leading to the ineffectiveness of 3D substrate on restraining the dendrites.To address these issues, 2D surface coatings have been introduced into the 3D porous skeletons with good zinc affinity to reinforce the electrode-electrolyte interfacial stability.For instance, an In@Zn@In triplex layer on a Cu mesh has been demonstrated to be effective for homogenized Zn nucleation. 33y constructing in situ-formed ZnF 2 -rich SEI layer on the MXene/GO aerogel to isolate Zn from electrolyte, the inactive byproducts Zn(OH) 4  2− and side reactions could be eliminated. 30Interestingly, these introduced coating layers are very thin (<100 nm); the thick coating on the 3D host results in high interfacial resistance and reduced energy density.Nanoscale interfacial modifications can decrease the resistance at the electrode-electrolyte interface, resulting in rapid kinetic transport and enabling high-energy Zn anode that contribute to extended cycling life. 34n this work, we develop a nanoscale 2D inorganicorganic composite interlayer (α-In 2 Se 3 -Nafion, denoted as ISNF) on the 3D porous CNTs substrate via complexing exfoliated α-In 2 Se 3 nanosheets with Nafion.The firstprinciples calculation results demonstrate that α-In 2 Se 3 nanosheets embrace lower nucleation barrier and strong affinity to Zn 2+ , which is capable of optimizing Zn nucleation behavior.Nafion cannot only shield anions and free H 2 O molecules from undesirable side reactions through its hydrophilic and hydrophobic regions, but also alleviate the volume changes during Zn stripping/plating owing to its high mechanical flexibility.Additionally, the framework of 3D CNTs with large surface area can afford sufficient free space to accommodate large quantities of Zn and homogenize the electric field distribution, ensuring the formation of dendrite-free electrodes.Taking these advantages, the composite Zn anode with ISNF@CNTs host (denoted as ISNF@CNTs@Zn) shows the voltage hysteresis of about 5 mV over 1500 h at 0.5 mA cm −2 .Furthermore, a flexible pouch cell with ISNF@CNTs@Zn anode and MnO 2 cathode delivers a good capacity retention of 86.5% after 1300 cycles, and admirable mechanical flexibility with negligible capacity fading under various deformations.

RESULTS AND DISCUSSION
α-In 2 Se 3 , a typical van der Waals layered material, has weak interlayer interaction that enables each individual layer to act independently, which gives rise to scalable preparation of large-area α-In 2 Se 3 thin film. 35,36  According to the previous reports, 35,37 the possible mechanism of the electrochemical process can be expressed as follows: The exfoliated α-In 2 Se 3 solution can be sufficiently mixed with Nafion, which forms a stable and easy-tohandle brownish yellow ISNF ink solution.With the assistance of simple spray-coating technology, the uniform ISNF film can be introduced onto the surface of homemade CNTs film synthesized, as in our previous study (Figure S1). 38,39e detailed morphological transformation of bulk α-In 2 Se 3 after exfoliation was then studied by transmission electron microscopy (TEM).As presented in Figure 1B, the bulk α-In 2 Se 3 shows an irregular particle morphology, while the exfoliated α-In 2 Se 3 has a 2D thin-sheet structure with a lateral size of 200-800 nm (Figure S2).The high-resolution TEM (HRTEM) image and corresponding selective area electron diffraction (SAED) patterns clearly reveal typical lattice spacings of 0.209 and 0.119 nm, which correspond to the (0111) and (0024) planes of α-In 2 Se 3 nanosheets (Figure 1C), respectively.Atomic force microscopy (AFM) image in Figure 1D indicates that these exfoliated α-In 2 Se 3 nanosheets have a thickness distribution of 4-5 nm, further confirming the successful exfoliation of bulk α-In 2 Se 3 .The energy-dispersive spectroscopy (EDS) elemental mapping further evidences the homogeneous distribution of In and Se elements in the exfoliated α-In 2 Se 3 nanosheets (Figure S3).The exfoliated α-In 2 Se 3 nanosheets covered on the surface of CNTs film show a uniform morphology, with a thickness of approximately 30-40 nm is achieved (Figure 1E and Figure S4).The inorganic α-In 2 Se 3 layer, however, is susceptible to cracking due to its high brittleness during the repeated Zn plating/stripping processes, ultimately compromising the protection effect of interfacial layer (Figure S5).To further reinforce the adhesion between α-In 2 Se 3 nanosheets and CNTs film, Nafion was mixed with α-In 2 Se 3 to form an ISNF protective layer with an average thickness of approximately 60 nm on the surface of CNTs film (Figure 1F,G and Figure S6).X-ray diffraction (XRD) measurement was then employed to investigate the phase structure of α-In 2 Se 3 @CNTs and ISNF@CNTs.As presented in Figure 1H, one peak at 18.5 • can be clearly observed in α-In 2 Se 3 @CNTs, which is indexed to the (006) plane of α-In 2 Se 3 (JCPDS: 34-0455), further confirming the successful formation of α-In 2 Se 3 protective layer on the CNTs film (Figure S7).However, the diffraction peak of (006) for ISNF@CNTs becomes weak owing to the shielding effect of Nafion.The full coverage of ISNF onto the CNTs film is highly expected to guide a dendrite-free Zn deposition (Figures S8 and S9).Besides, the surface chemical composition of ISNF@CNTs was investigated by X-ray photoelectron spectroscopy (XPS).][42] To clarify the zincophilic properties of each component in ISNF@CNTs, the density functional theory (DFT) calculation was conducted, and the corresponding optimized crystal structures are presented in Figure 2A,B.Impressively, the binding energy of a Zn atom on the surface of α-In 2 Se 3 can reach −0.38 eV, which is remarkably lower than that of CNTs (−0.04 eV) and Nafion (−0.06 eV) (Figures S12).After Nafion doping, the binding energy of a Zn atom on the ISNF surface is further decreased to −0.80 eV due to the synergistic effect of α-In 2 Se 3 and Nafion.Meanwhile, the local charge density between the Zn atom and ISNF is also increased, showing that there is a strong interaction between ISNF and Zn.These calculations indicate that the ISNF protective layer may effectively enhance the homogeneity of Zn nucleation distribution.The hydrophilicity of the inorganic-organic protective layer with respect to the electrolyte was investigated by monitoring the dynamic contact angle of CNTs with/without the ISNF coating layer at 25 • C. As shown in Figure 2C,D, the initial contact angle of the untreated CNTs film is approximately 94 • , and it is slightly reduced to 89 • in the following 5 min (Figure S13).After 20 min, a large contact angle of 57 • is still observed, revealing the limited hydrophilicity of CNTs film in aqueous electrolyte.In contrast, the initial contact angle of ISNF@CNTs film is found to be 67 • , significantly lower than that of the CNTs film.In the ensuing 20 min, it rapidly decreases to 42.3 • , demonstrating that the artificial ISNF@CNTs film remarkably improves the hydrophilicity.][45] The nucleation overpotential of CNTs@Zn, Nafion@CNTs@Zn, α-In 2 Se 3 @CNTs@Zn, and ISNF@CNTs@Zn electrodes was measured against Zn at 1 mA cm −2 (Figure 2E).It is found that the ISNF@CNTs@Zn electrode shows a nucleation overpotential of 30 mV, which is lower than that of CNTs@Zn, Nafion@CNTs@Zn, and α-In 2 Se 3 @CNTs@Zn electrodes.The lowered polarization of ISNF@CNTs@Zn electrode reveals that the nucleation and growth of Zn is more conducive to proceed on the ISNF layer surface than on the bare CNTs@Zn surface.The ISNF@CNTs@Zn electrode exhibits a smaller charge-transfer resistance compared with the other three electrodes, thus providing a higher ionic conductivity to ensure faster Zn deposition kinetics on the ISNF@CNTs@Zn surface (Figure 2F).The enhanced charge-transfer capability of ISNF@CNTs@Zn is also demonstrated in terms of σ ionic (Figure 2G).σ ionic of ISNF@CNTs@Zn is approximately 77 mS cm −1 , much higher than approximately 43 mS cm −1 of α-In 2 Se 3 @CNTs@Zn, approximately 26 mS cm −1 of Nafion@CNTs@Zn, and approximately 10 mS cm −1 of CNTs@Zn.Moreover, the hydrogen evolution and Zn corrosion, which are two undesired side reactions originating from water decomposition and oxygen dissolved in the electrolyte, can be restrained in ISNF@CNTs@Zn.To support this conclusion, the linear sweep voltammetry (LSV), Tafel plots, and linear polarization (LP) measurements of CNTs@Zn, Nafion@CNTs@Zn, α-In 2 Se 3 @CNTs@Zn, and ISNF@CNTs@Zn electrodes were conducted using a three-electrode system.As presented in LSV curves (Figure 2H), the onset potential of different electrodes is in the order of CNTs@Zn (−1.46 V) > Nafion@CNTs@Zn (−1.67 V) > α-In 2 Se 3 @CNTs@Zn (−1.78 V) > ISNF@CNTs@Zn (−1.87 V), which reveals the reduced hydrogen evolution activity of ISNF@CNTs@Zn.The Tafel slope of ISNF@CNTs@Zn (523 mV dec −1 ) is much higher than those of CNTs@Zn (182 mV dec −1 ), Nafion@CNTs@Zn (218 mV dec −1 ), and α-In 2 Se 3 @CNTs@Zn (504 mV dec −1 ) (Figure S14).
To demonstrate the superiority of ISNF@CNTs host, the Zn-plating morphologies on different hosts were studied in Figure 3A-F.The non-uniform Zn nanoparticles were formed on the CNTs surface after the deposition of 2 F I G U R E 3 Exploration of Zn plating/stripping behavior on CNTs and ISNF@CNTs electrodes.Top-view SEM images of (A) CNTs and (D) ISNF@CNTs electrodes after plating Zn at 2 mA cm −2 with a capacity of 2 mAh cm −2 .Top-view SEM images of (B) CNTs and (E) ISNF@CNTs electrodes after plating Zn at 5 mA cm −2 with a capacity of 5 mAh cm −2 .Side-view SEM images of (C) CNTs and (F) ISNF@CNTs electrodes after plating Zn at 5 mA cm −2 with a capacity of 5 mAh cm −2 .High-resolution XPS spectra of ISNF@CNTs after plating Zn with different etching depths: (G) Zn 2p spectrum, (H) In 3d spectrum, and (I) Se 3d spectrum.Simulated electric field distributions on (J) CNTs@Zn and (K) ISNF@CNTs@Zn electrodes.(L) Schematic illustrations of the Zn deposition on CNTs and ISNF@CNTs electrodes.and 5 mAh cm −2 , and there are several conspicuous protuberances detected, indicative of the inhomogeneous Zn plating (Figure 3A-C).Unsurprisingly, some obvious Zn protrusions are found on the surface of the Nafion@CNTs and α-In 2 Se 3 @CNTs even after 5 mAh cm −2 of Zn plating, leading to dendrite Zn plating (Figures S15 and S16).However, the ISNF@CNTs can well preserve the initial smooth morphology after Zn plating (Figure 3D-F), reflecting that the ISNF protective layer can guide uniform Zn plating underneath the film, which is further confirmed by XPS with different etching depths (Figure 3G-I).7][48] With increasing etching depths, the metallic Zn peaks in ISNF@CNTs gradually increase, while the peak intensities of In 3+ and Se 2− clearly decrease, suggesting that the Zn plating grows beneath the surface of ISNF.To further analyze the Zn plating/stripping behavior of ISNF@CNTs host, the simulation of electric field distribution for CNTs@Zn and ISNF@CNTs@Zn electrodes was performed in Figure 3J,K.Owing to the uneven Zn nucleation on CNTs, the subsequent Zn plating process is found to have inhomogeneous electric field distribution, along with a higher charge region near the Zn nuclei.The distinct field intensity gradient enables more Zn 2+ flux to be concentrated at the individual nucleation sites.During cycling, the Zn 2+ ions preferentially deposit on these protuberance sites, which brings about uncontrolled Zn dendrites.Fortunately, after the introduction of the ISNF@CNTs host, the electric field streamlines become more homogeneous, and such unique characteristics can provide adequate zincophilic nucleation sites.The homogeneously distributed electric field effectively eliminates the "tip effect" occurred on the CNTs counterpart, guaranteeing more uniform Zn 2+ adsorption in the electrode.0][51] These results are consistent with those observed using in situ optical microscopy at a current density of 2 mA cm −2 (Figure S17).As expected, a non-uniform Zn deposition with visible bulges appears on the surface of CNTs hosts as soon as 120 s.These protrusions remarkably disrupt the local electrical distribution, resulting in the successive erratic growth of dendrites that may connect the electrodes and cause short-circuiting.However, as the plating time increases, the INSF@CNTs electrode still exhibits uniform Zn deposition without any surface protrusions, underscoring the effectiveness of the ISNF protective layer in facilitating homogeneous Zn deposition and enhancing cycling stability.According to the above theoretical calculations and experimental results, the schematic illustrations of Zn deposition on CNTs and ISNF@CNTs have been proposed, as shown in Figure 3L.When CNTs films are employed as the Zn plating hosts, Zn 2+ ions tend to be plated onto these limited nucleation sites during the electrochemical deposition process, which tends to cause Zn dendrite formation.However, when a 3D ISNF@CNTs serve as the current collector, the final morphology can be obviously improved.The ISNF protective layer on the surface of CNTs can effectively prevent anion and free water molecule from reacting with the active Zn sources, subsequently inhibiting the generation of undesirable byproducts and hydrogen evolution.
during the repeated Zn plating/stripping process.In contrast, the ISNF@CNTs@Zn shows a smooth and homogeneous topography without any obvious large grains even after prolonged cycling, which can be ascribed to the uniform deposition of Zn and the anticorrosive behavior provided by the ISNF@CNTs hosts.The Coulombic efficiency (CE) of Zn plating/stripping was investigated based on asymmetric cells (Figure 4G).It is observed that the ISNF@CNTs||Zn battery demonstrates an extended cycle life of 1200 cycles, maintaining an average CE of approximately 99.6%, whereas the CNTs||Zn displays significant degradation after only 54 cycles.Figure 4H compares the rate performance of the CNTs@Zn, Nafion@CNTs@Zn, α-In 2 Se 3 @CNTs@Zn, and ISNF@CNTs@Zn electrodes at different current densities, in which the ISNF@CNTs@Zn electrode always presents lowest overpotential, especially at high current densities.It is evidently clear that when the current density is turned back to 0.5 mA cm −2 , the overpotential remains unchanged for ISNF@CNTs@Zn, whereas it rises by approximately 55.6% for α-In 2 Se 3 @CNTs@Zn, suggesting low polarization and improved structural stability of ISNF@CNTs@Zn electrode.
serves as the electrolyte.Figure 5A displays the cyclic voltammetry (CV) profiles of the CNTs@Zn||MnO 2 and ISNF@CNTs@Zn||MnO 2 batteries.3][64][65] It is worth noting that the ISNF@CNTs@Zn||MnO 2 battery has a lower overpotential gap than that of CNTs@Zn||MnO 2 one, indicating its smaller polarization and faster charge transfer rate.This can be further affirmed by its smaller transfer resistance revealed by ISNF@CNTs@Zn||MnO 2 battery in electrochemical impedance spectroscopy (EIS) measurements in Figure 5B and Figure S26.The charge/discharge properties of the two batteries were also compared (Figure 5C).The ISNF@CNTs@Zn||MnO 2 battery delivers a discharge capacity of 253.2 mAh g −1 at 0.5 A g −1 , while the CNTs@Zn||MnO 2 battery shows an inferior capacity of 211.8 mAh g −1 under the same operating conditions.Even at a high current density of 5 A g −1 , the ISNF@CNTs@Zn||MnO 2 battery still exhibits reversible capacity of 118.3 mAh g −1 , which is much better than that of CNTs@Zn||MnO 2 counterpart (69.2 mAh g −1 ).In addition, it shows superior cycling stability (Figure 5E-G).After 2500 cycles of charge/discharge at 3 A g −1 , the CNTs@Zn||MnO 2 battery decays from a capacity of 141.4 to 77.1 mAh g −1 , corresponding to a capacity retention rate of 54.5%.However, the ISNF@CNTs@Zn||MnO 2 battery delivers a capacity of 142.2 mAh g −1 , and maintains a capacity retention rate of 80.1% after 2500 cycles (Figure S27).Besides, the open circuit voltage (OCV) decay of the fully charge CNTs@Zn||MnO 2 and ISNF@CNTs@Zn||MnO 2 batteries was monitored during 48 h of storage (Figure 5H).The batteries were cycled 10 times at 0.3 A g −1 before testing.The OCV of the CNTs@Zn||MnO 2 battery decreases to 1.43 V after 48 h.In the following discharge capacity, it manifests only 59% of the initial capacity owing to the rapid corrosion of CNTs@Zn electrode.In contrast, the ISNF@CNTs@Zn||MnO 2 battery maintains the OCV at 1.49 V after 48 h.Subsequently, the battery yields 90% of its original capacity.All these findings reveal that the ISNF@CNTs@Zn electrode is a promising host for improving the electrochemical properties of Zn anode.
To further prove the superiority of the ISNF@CNTs@Zn electrode for wearable electrochemical energy storage systems, a flexible quasi-solid-state ZIB was assembled by employing PVA/ZnSO 4 /MnSO 4 /Na 2 SO 4 as gel electrolyte (Figure 6A).Interestingly, this quasi-solid-state ZIB demonstrates similar charge/discharge profiles to the one in 3 M ZnSO 4 /MnSO 4 /Na 2 SO 4 aqueous electrolyte, which suggests an effective infiltration of PVAbased gel electrolyte into the entire device (Figure S28). Figure 6B shows the capacities of the quasi-solid-state ISNF@CNTs@Zn||MnO 2 battery at various current densities (0.5, 1, 2, 3, and 5 A g −1 ), which is superior to many reported ZIBs using MXene-rGO2, 66 α-MnO 2 , 67 ZnHCF, 68 γ-MnO 2 , 69 VO 2 @rGO, 70 δ-MnO 2 , 71 and ZnMn 1.68 O 4 72 as cathodes (Figure 6C).Benefiting from the satisfactory rate capacity, the quasi-solid-state ZIB with the ISNF@CNTs@Zn electrode outputs an energy density of 253.5 Wh kg −1 at the power density of 0.65 kW kg −1 (based on the mass of cathode active material).In addition, after over 1300 cycles at 3 A g −1 , this quasi-solidstate ISNF@CNTs@Zn||MnO 2 battery can still deliver the capacity retention of 86.5% (Figure 6D).The mechanical flexibility of ZIB plays a crucial role in the application of wearable electronics.As expected, the as-fabricated ISNF@CNTs@Zn||MnO 2 battery maintains more than 85% capacity retention after being bent and twisted and still survive over 350 cycles, showing its wonderful flexibility (Figure 6E,F).Last but not the least, the quasi-solid-state ZIB is still capable of powering an electronic watch after several times bending (Figure 6G).These results clearly demonstrate that our ISNF@CNTs@Zn electrodes are highly suitable for the design of flexible high-performance ZIBs.

CONCLUSION
In summary, a nanosalce inorganic/organic film (ISNF) has been deposited on CNTs and used as a Zn plating/stripping host to accomplish dendrite-free ISNF@CNTs@Zn anode.Based on the integrated evidence from experimental observations and DFT calculations, it has been found that the unique ISNF protective layer can significantly suppress the intricate side reactions of the electrolyte and inhibit the formation of Zn dendrites, thus promoting uniform Zn plating/stripping with high reversibility.A freestanding 2D/3D host with high electronic conductivity can favor rapid electrolyte penetration and accelerated electron/ion transport.In comparison with CNTs@Zn, Nafion@CNTs@Zn, and α-In 2 Se 3 @CNTs@Zn, the ISNF@CNTs@Zn symmetric cell demonstrates a superior long-term cycle life over 1500 h with a lower overpotential.When employing a MnO 2 electrode as cathode, both the aqueous full cells and flexible quasi-solid-state cell exhibit low polarization, excellent rate capability, and wonderful durability.Most surprisingly, the ISNF@CNTs@Zn electrode in the flexible quasi-solid-state cell shows almost no change in capacity after bending and twisting.This work is expected to provide a feasible strategy for constructing 3D hosts to achieve high-stable dendrite-free Zn anode toward next-generation advanced batteries.

Preparation of α-In 2 Se 3 nanosheets
α-In 2 Se 3 nanosheets were synthesized via a facile electrochemical intercalation method using a two-electrode configuration. 35In details, a thin piece of α-In 2 Se 3 and graphite rod were chosen as the cathode and anode, respectively.The α-In 2 Se 3 piece was prepared by using a conductive copper mesh to wrap bulk α-In 2 Se 3 powder.The electrolyte was made of tetraheptylammonium bromide and acetonitrile (5 mg mL −1 , 80 mL).The applied voltage was set to be 20-25 V.After complete delamination (2 h), the exfoliated α-In 2 Se 3 was collected by centrifuge and washed with absolute ethanol.To further enlarge the interlayer distance, the as-intercalated sample was sonicated in 50 mL 0.2 M PVP/DMF solution under ice bath for 2 h.Subsequently, the reddish-brown dispersion was thoroughly washed with isopropanol (IPA) to eliminate excessive PVP and then dispersed into IPA solution under mild sonication.Finally, a low-concentration (approximately 2 mg mL −1 ) colloidal solution of α-In 2 Se 3 nanosheets was obtained by getting rid of the thick-layered α-In 2 Se 3 flakes after centrifugation at 9000 rpm for 10 min.

Preparation of ISNF@CNTs
A dispersion solution of ISNF was prepared through mixing 100 μL Nafion membrane solution (5%) with 10 mL α-In 2 Se 3 solution in an ice bath.Afterwards, the ISNF solution was sprayed on the CNTs films, which were obtained via our previous method, 38 to fabricate ISNF@CNTs electrodes.The mass loading of CNTs and ISNF in ISNF@CNTs was about 1 and 0.15 mg cm −2 , respectively.Nafion@CNTs and α-In 2 Se 3 @CNTs electrodes were prepared with the same procedure without Nafion or α-In 2 Se 3 solution.

Preparation of ISNF@CNTs@Zn
ISNF@CNTs@Zn electrode was synthesized by a simple electrodeposition method.A platinum sheet (30 × 10 × 0.1 mm) was placed as the counter electrode and a Ag/AgCl (saturated KCl) electrode was served as the reference electrode. 73The electrolyte used for deposition was composed of 0.35 M ZnSO 4 and 0.57 M Na 3 C 6 H 5 O 7 ⋅2H 2 O solution.Electrodeposition of ISNF@CNTs@Zn was performed at a current density of 10 mA cm −2 for 10 min and 20 mA cm −2 for 5 min at room temperature (the mass loading of Zn active materials in whole electrode was estimated to be approximately 5-6 mg cm −2 ).

Preparation of MnO 2 cathode
α-MnO 2 nanorods were prepared according to a modified hydrothermal method. 74Briefly, 2 mL of 0.3 M HCl was added into 10 mL of 0.03 M MnSO 4 under stirring.Subsequently, 20 mL of 0.1 M KMnO 4 was slowly added into the above solution.After continuous stirring for 1 h, the mixture was transferred into a Teflon-lined autoclave and heated at 120 • C for 12 h.Finally, the black precipitate was collected through centrifugation in vacuum at 70 • C for 24 h.The MnO 2 cathode was first fabricated by mixing α-MnO 2 (70%), carbon black (20%), and PVDF (10%) in NMP solution.Then, the obtained slurry was cast on the surface of carbon paper and dried at 70 • C for 48 h under vacuum.The areal mass loading for the MnO 2 cathode was approximately 1-2 mg cm −2 .

Assembly of flexible quasi-solid-state ZIBs
The flexible quasi-solid-state ZIBs were assembled by employing ISNF@CNTs@Zn as anode, MnO 2 electrode as cathode, and PVA/ZnSO 4 /MnSO 4 /Na 2 SO 4 gel as electrolyte.The PVA-based gel electrolyte was fabricated by dissolving 2 g of PVA, 9.6 g of ZnSO 4 , 0.15 g of Mn 2 SO 4 , and 0.14 g of Na 2 SO 4 in 20 mL aqueous solutions at approximately 80 • C-90 • C.Then, both MnO 2 and ISNF@CNTs@Zn electrodes were immersed in PVA-based gel electrolyte for 3 min and dried under vacuum at 40 • C for 1 h.Finally, the two electrodes were face-to-face assembled, and then the flexible device was placed overnight before use.

Simulation of electric field contribution
A simplified 2D electrodeposition model was constructed to simulate the proportional schematics of electric field distribution during cycling based on COMSOL Multiphysics software. 75In this model, the length was set to 4 μm.The ionic conductivity of 3 M ZnSO 4 aqueous electrolyte was approximately 5 S m −1 .For the ISNF@CNTs@Zn electrode, the single ISNF@CNTs was represented by rectangles with a width of 400 nm and a height of 500 nm.The gap between ISNF@CNTs nanosheets was set to 40 nm.The electrical conductivity of ISNF and Zn metal was measured to be 6.96 × 10 −5 and 1.67 × 10 7 S m −1 , respectively.The voltage excitation between the electrolyte and anode side was 500 mV.

DFT calculation
Density functional theory (DFT) calculations were conducted using projector-augmented wave (PAW) method, as implemented in the Vienna ab initio simulation package (VASP). 76The exchange-correlation interaction was investigated by employing Perdew-Burke-Ernzerhof (PBE) type generalized gradient approximation (GGA). 77In this simulation, all calculations used an energy cutoff of 450 eV and gamma centered 2 × 3 × 1 k-points mesh.To further describe the adsorption behavior between deposited Zn atom and electrode, a vacuum layer of 15 Å was applied.
The residual forces were controlled to below 0.05 eV Å −1 after relaxation.

Characterizations
The morphology images for samples were characterized by scanning electron microscopy (SEM, JEOL JSM-7001F) with an accelerating voltage 5.0 kV.The microstructures of the electrodes were tested using transmission electron microscopy (TEM, JEOL JEM-F200).The crystalline structure of the materials was analyzed by a Bruker D8 Advance x-ray diffractometer with Cu Kα radiation (λ = 0.15406 nm) operating at 30 kV and 10 Ma.The Fourier transform infrared (FTIR) spectra were collected on a PerkinElmer Spectrum One spectrometer.Raman spectra were recorded at room temperature using a Renishaw InVia Raman microscope with a 633 nm laser.The wettability of electrodes was conducted on a contact anglemeasuring device (Kruss, DSA30).The surface chemistry of electrodes was analyzed by x-ray photoelectron spectroscopy (XPS) using an ESCALAB 250Xi (Thermo Fisher) spectrometer.Atomic force microscopy (AFM) spectra were carried out on a Bruker Dimension ICON microscope with a SCANASYST-AIR probe.The H 2 production was quantitatively measured using a Shiweipuxin GC7806 gas chromatograph.The morphological changes during Zn plating were observed using the Spotlight 400 FT-IR imaging system.

Electrochemical characterization
The Zn plating/stripping behaviors were studied in an asymmetric cell system by using ISNF@CNTs, α-In 2 Se 3 @CNTs, Nafion@CNTs, or CNTs films as cathode materials and Zn foil as anode.The symmetric cells were assembled in air by employing two identical electrodes as counter and working electrodes, respectively.CV, LSV, LP curves, and electrochemical impedance spectroscopy (EIS) for as-prepared samples were carried out on a CHI 760E electrochemical working station.The galvanostatic charge/discharge (GCD) measurements were performed on a multichannel LAND battery testing system (CT3001A).

A
C K N O W L E D G M E N T S This work was supported by the Natural Science Foundation for Young Scientists of Henan Province (grant number: 202300410071), the Key Research Project of Henan Provincial Higher Education (grant number: 21A140007), and the National Natural Science Foundation of China (grant numbers: 62174049, 52003073, and 52102285).C O N F L I C T O F I N T E R E S T S TAT E M E N T The authors declare no conflict of interest.O R C I D Chong Chen https://orcid.org/0000-0003-2940-2030R E F E R E N C E S