English

• Na-ion batteries (NIBs) are recognized as one of the most promising energy storage technologies for green-grid systems due to the high abundance of sodium resources [1-4]. However, compared to the advanced lithium-ion batteries (LIBs) (approaching 300 Wh/kg), the energy density of NIBs needs to be further increased [5]. One of the main factors that affect their development is the lack of affordable and high-performance anode materials. Conversion-type anode materials (M_aX_b; M = Ni, Co, Fe, Mn, Cu, none, etc.; X = O, S, Se, Te, P, N, etc.) undergo multi-electron reactions, presenting high theoretical capacity [6-10]. Meanwhile, they are usually low-cost, environmentally friendly, and non-toxicity. Hence, conversion-type anode materials are considered very promising for obtaining NIBs with high energy density. One of the key obstacles restricting their developments is the poor cycling stability. During the intercalation of Na⁺ ions, the atoms in M_aX_b structure have been rearranged, and M_aX_b is converted into nano-sized M particles and Na_cX phases [7,11]. The above molecular-level reorganization process results in huge volume changes and finally severe pulverization of the active materials, leading to an inferior electrochemical performance of M_aX_b. In response to the above problems, the conventional optimization strategies include nano-sized structures design, doping and adding electronic conductivity additives [6,12]. Spatial confinement effect enabled by constructing a fast electronic/ionic network around the M_aX_b materials is promising for achieving a long cycling life. This network could help to withstand volume expansion of the sodiated phases while improving electron/ion transport kinetics.

  When it comes to constructing a robust 3D electronic/ionic network, constructing 3D graphene framework (GF) as a conductive network is an effective strategy to improve the electrochemical performance of sulfur-based cathode [13]. The 3DGF is constructed by conjugated and interlocked graphene nanosheets, forming a continuous and hierarchical porous network. It can withstand the volume expansion of sodiated phases and provide an efficient ion transport path. Besides, the 3DGF displays a high conductivity of up to ~1400 S/m, which benefits to improve the electronic conductivity of composite materials [13]. Moreover, the strong mechanical strength of 3DGF can make a freestanding electrode without inactive binders. Considering the above advantages, 3DGF can be applied as a robust ionic/electronic conductive network in conversion-type anode materials.

  Fe₂O₃ anode presents a theoretical capacity of up to 1007 mAh/g and is one of the most promising conversion-type anodes of NIBs [14]. Herein, a self-supporting 3D graphene sponge decorated with Fe₂O₃ nanocubes (rGO@Fe₂O₃) was designed. It should be emphasized that the carbon composite is a promising strategy to optimize properties [15-18]. In the reported work, graphene mainly presents a two-dimensional (2D) sheet structure. However, we constructed a three-dimensional (3D) hierarchical porous network structure by cross-linking the 2D graphene sheet, in which the Fe₂O₃ nanocubes are encapsulated by graphene. The 3DGF provides a high-porosity structure, and Fe₂O₃ nanocubes are well dispersed in 3DGF without agglomeration. The 3DFG can not only act as a structural framework to buffer the volume expansion of Fe₂O₃ nanocubes but also serve as an ionic/electronic transport network. In particular, the rGO@Fe₂O₃ sponge can be directly acted as a self-supporting electrode without other inactive components such as binders and conductive additives. Subsequently, the electrochemical performance test, morphology evolution, and reaction kinetics analysis of the materials were carried out. It is found that sodiated Fe₂O₃ nanocubes in rGO@Fe₂O₃ sponge have not yet pulverized. The rGO@Fe₂O₃ sponge delivers excellent electrochemical performance. Finally, the assembled full NIBs display high power and energy density. This work unveils the potential of achieving NIBs with exceptional longevity through constructing a 3D ionic/electronic conductive network decorated with a nanosized conversion-type anode.

  The high-porosity rGO@Fe₂O₃ sponge was synthesized by self-assembly and calcination processes (Fig. 1a). First, graphene oxides decorated with Fe₂O₃ nanocubes (GO@Fe₂O₃) hydrogel were obtained by self-assembly under hydrothermal conditions. Then, the 3D rGO@Fe₂O₃ sponge was prepared by further annealing in the N₂ atmosphere. The pure Fe₂O₃ nanoparticles were also synthesized for comparison. XRD patterns of the Fe₂O₃ and rGO@Fe₂O₃ are shown in Fig. S1a (Supporting information), and the phase is α-Fe₂O₃ (ICCD 01–084–0308). In the Raman spectra of rGO@Fe₂O₃, all peaks are assigned to rGO and Fe₂O₃ (Fig. S1b in Supporting information). The obvious peaks between 200 and 300 cm⁻¹ are derived from the Fe₂O₃ [19]. The graphitic carbon (G band, 1590 cm⁻¹) and the disordered carbon (D band 1351 cm⁻¹) are also observed [20]. The measured I_D/I_G of around 1.08 indicates the presence of defects in rGO which help nanoparticles adhere to the surface of rGO [20]. The XPS survey spectrum displays Fe, O, C and N elements in the rGO@Fe₂O₃ sponge (Fig. S2a in Supporting information). Additionally, the Fe 2p XPS spectrum can be fitted into an one doublet Fe 2p_1/2 (724.9 eV) and Fe 2p_3/2 (711.3 eV) [15,21], indicating the existence of Fe₂O₃, coincident with the result of XRD pattern (Fig. S2b in Supporting information). In addition, the XPS spectrum of C 1 s can be fitted into three peaks at 284.7, 285.8, and 288.9 eV, ascribing to the C-C, O-C=O, and C-O (Fig. S2c in Supporting information). The results indicate the existence of oxygen-containing groups on the surface of rGO which could anchor Fe₂O₃ nanocubes by the Fe-O-C bond, restricting its growth and making the nanoparticles dispersed evenly.

  Figure 1

  

  Figure 1.  Schematic of the fabrication and structural characterization of rGO@Fe₂O₃ sponge. (a) Schematic illustration for the synthesis of rGO@Fe₂O₃ sponge. (b, c) SEM and TEM images of rGO@Fe₂O₃ sponge. (d, e) Nitrogen adsorption and desorption isotherm and pore-size distribution of rGO@Fe₂O₃ sponge.

  DownLoad: Full-Size Img PowerPoint

  The morphology of Fe₂O₃ and rGO@Fe₂O₃ sponge was further investigated. The average size of pure Fe₂O₃ nanocubes is about 140 nm (Fig. S3 in Supporting information). As for the rGO@Fe₂O₃ sponge, it is found that Fe₂O₃ nanocubes are dispersed on the rGO surface without agglomeration (Figs. 1b and c), that is conducive to expose more active sites and further obtain excellent electrochemical performance. The Fe₂O₃ nanocubes exhibit an average size of about 100 nm. The high-resolution TEM (HRTEM) image reveals that Fe₂O₃ nanocubes are in close contact with the rGO sheets with a thickness of less than 5 nm (Fig. S4a in Supporting information), beneficial to fast electron transport. The distinct lattice fringes present a d-spacing of 0.36 nm and 0.27 nm, matching to the (012) plane and (104) plane of the α-Fe₂O₃ (Fig. S4b in Supporting information). The diffraction rings mean the polycrystalline characteristic (inset Fig. S4b). Besides, the distribution of Fe, O, C, and N elements is in good agreement with the rGO@Fe₂O₃ morphology (Fig. S5 in Supporting information). Meanwhile, the nitrogen element is attributed to nitrogen doping in rGO, consistent with the XPS analysis results (Fig. S2a). The 3D porous structures of the rGO@Fe₂O₃ are investigated by nitrogen adsorption-desorption isotherms (Fig. 1d) and pore-size distributions (Fig. 1e). The Brunauer–Emmett–Teller specific surface area of rGO@Fe₂O₃ is 31.87 m²/g. The pores of rGO@Fe₂O₃ range mostly between 2 nm and 100 nm based on the Barrett-Joyner-Halenda model. Interestingly, the well-defined mesopores (~3 nm/40 nm) and macropores (~60 nm) were obtained, presenting a hierarchical porous network, favorable for electrolyte accessibility and rapid Na-ion diffusion (Fig. 1e).

  To evaluate the Na⁺-ion storage performance, the electrode slices were first prepared, with sodium metal as a counter electrode. The rGO@Fe₂O₃ sponge can be directly used as electrode slices without adding additional binders and conductive carbon black (Fig. S6 in Supporting information), further achieving batteries with higher gravimetric energy density. Based on the CHN elemental analysis, the carbon content of rGO@Fe₂O₃ is around 56% (Table S1 in Supporting information). The gravimetric capacity of the rGO@Fe₂O₃ sponge is calculated based on the quality of the entire electrode. The initial discharge/charge capacities are 1683/859 mAh/g with a first Coulombic efficiency of 51% at 0.1 A/g in the voltage range of 0.01–3 V vs. Na⁺/Na (Fig. 2a). The capacity loss may ascribe to the irreversible consumption of sodium ions by the functional groups on the graphene sheet and the formation of solid electrolyte interface (SEI) film. After initial cycle, rGO@Fe₂O₃ electrode exhibits high reversibility.

  Figure 2

  

  Figure 2.  Electrochemical performances of rGO@Fe₂O₃ sponge and Fe₂O₃. (a) The galvanostatic charge-discharge curves of rGO@Fe₂O₃ sponge at 0.1 A/g. (b, c) CV curves of rGO@Fe₂O₃ and Fe₂O₃ at 0.2 mV/s at first three cycles. (d) Long-term cycles at 1 A/g of rGO@Fe₂O₃. (e) Cycling performance comparison of current work to the state of the art. (f) Rate performance of rGO@Fe₂O₃ and Fe₂O₃ at different current densities. (g) Rate performance comparison of current work to the state of the art.

  DownLoad: Full-Size Img PowerPoint

  Figs. 2b and c display the cyclic voltammetry (CV) curves. The irreversible cathodic peaks at the initial cycle correspond to the SEI formation. Compared with Fe₂O₃ anode, the first-loop CV curves of rGO@Fe₂O₃ have more irreversible peaks, which may be due to the irreversible redox reaction of the functional groups on the surface of rGO, and this may also be the reason for the low initial Coulomb efficiency. In the second cycle, there are peak 1 and 2 (0.76 V, 0.01 V) on the reduction side and peak 3 and 4 (0.3 V, 1.5 V) on the oxidation side. The peak 1 indicates the Na⁺ insertion process into Fe₂O₃, corresponding to the Na₂Fe₂O₃ formation [11]. The peak 2 can be attributed to the phase conversion from Na_xFe₂O₃ to Na₂O and Fe. During Na⁺ extraction, the peak 3 is assigned to the oxidation of Fe, resulting in the formation of Na₂Fe₂O₃. The peak 4 at 1.5 vs. Na⁺/Na is attributed to the desodiation of Na₂Fe₂O₃. Consequently, the reversible conversion reaction can be described as followed: Fe₂O₃ + 6Na⁺+ 6e⁻↔ 2Fe + 3Na₂O [11,22]. The CV curves of pure Fe₂O₃ anode show a similar reaction process. The critical point is that the two anodic peaks (peak 3’ and peak 4’) of the pure Fe₂O₃ anode shift toward higher potential side than that of the rGO@Fe₂O₃, resulting a larger overpotential. The lower overpotential of the rGO@Fe₂O₃ anode indicates that the 3D conductive network framework can effectively reduce its impedance of desodiation reaction and improve the reaction kinetics.

  The cycling performance of the rGO@Fe₂O₃ sponge was also evaluated (Fig. 2d). Compared with Fe₂O₃, the rGO@Fe₂O₃ sponge shows higher capacity. After 2000 cycles at 1 A/g, the discharge capacity of rGO@Fe₂O₃ sponge is 327 mAh/g, which maintains 63.4% of the original capacity. The specific capacity of rGO@Fe₂O₃ decreases within 100 cycles and then stabilizes, and the corresponding polarization also increases first and then remains unchanged (Fig. S7a in Supporting information). These results suggest that the fading of capacity may be related to the increase of polarization. In addition, the 3D network structure can provide abundant active interfacial sites that may decrease during the cycling process, leading to the capacity decay. Fig. 2f shows the rate capability of rGO@Fe₂O₃ sponge and Fe₂O₃ at various current densities. The reversible capacity of Fe₂O₃ only exhibits 50 mAh/g at 4 A/g. In contrast, the reversible capacity of rGO@Fe₂O₃ anode is 271 and 162 mAh/g at 4 A/g and 20 A/g. The corresponding galvanostatic charge-discharge curves maintain the inclined-line shape at different current densities, indicating a fast capacitive behavior (Fig. S7b in Supporting information). The comparisons of cycling performance and rate capacity between previous reports and current work are shown in Figs. 2e and g [14-16,20,21,23-42]. It demonstrates that the Fe₂O₃ anode in this work has an obvious improvement in rate and cycling performance with the help of a 3D electronic/ionic network. The optimized performance may be attributed to its unique structure. 3D rGO conductive framework with high porosity and resilience facilitates the transport of the electrons/ions and accommodates the volume expansion of the Fe₂O₃ nanocubes during the desodiation/sodiation.

  The kinetics process of Na⁺ ion diffusion was studied by the galvanostatic intermittent titration technique (GITT) method [43-45]. The electrode was activated after three cycles to reach the equilibrium state for the GITT test. The time required for rGO@Fe₂O₃ anode to complete GITT discharge or charge test is as long as 25 h, much higher than the result of pure Fe₂O₃ anode (10 h, Fig. 3a). The Na⁺ diffusion coefficients (D_Na) for rGO@Fe₂O₃ anode at the initial states of discharging and charging are initially high at about 9.5 × 10⁻⁸ and 3.6 × 10⁻⁸ cm²/s (Fig. 3b and Fig. S8 in Supporting information). The calculated average D_Na⁺ over the entire discharge or charge process are about 2.12 × 10⁻⁸ and 3.19 × 10⁻⁸ cm²/s, significantly higher than those of pure Fe₂O₃ (1.34 × 10⁻⁸ and 9.7 × 10⁻⁹ cm²/s). In the electrochemical impedance spectrum (EIS) measurements, the rGO@Fe₂O₃ anode at a fully charged state presents a smaller semicircle than that of Fe₂O₃ (Fig. 3c), indicating a smaller charge-transfer resistance. The fast Na⁺ diffusion kinetics and low charge-transfer resistance of the rGO@Fe₂O₃ anode could be attributed to a unique 3D ionic/electronic conductive network.

  Figure 3

  

  Figure 3.  Electrochemical kinetic analysis. (a) GITT profiles of discharge and charge processes of rGO@Fe₂O₃ and Fe₂O₃ measured after three cycles at the current density of 0.02 A/g. (b, c) Na⁺ diffusion coefficients as a function of the states of charging process and Nyquist plots of rGO@ Fe₂O₃ and Fe₂O₃ at fully charged state. (d) CV curves of rGO@Fe₂O₃ at the scan rates ranging from 0.2 mV/s to 1 mV/s. (e)The normalized peak currents versus scan rate to determine the b value for anodic and cathodic peaks of rGO@Fe₂O₃. (f) The proportion of diffusion−controlled and capacitive contributions to the total charge storage at different scant rates.

  DownLoad: Full-Size Img PowerPoint

  The CV measurements were also carried out to study reaction kinetics during sodiation/desodiation process. When increasing the scan rates, the cathodic peak of the rGO@Fe₂O₃ electrode still exists and only slightly shifts (Fig. 3d and Fig. S9a in Supporting information). However, the cathodic peak of the Fe₂O₃ electrode disappeared (Figs. S9b and c in Supporting information). This result shows that rGO@Fe₂O₃ exhibits good charge transfer and ion diffusion ability during fast charge and discharge, effectively reducing the polarization of the electrode and improving its rate capacity. Then, as proposed by Dunn et al., a semi-quantitative analysis is conducted. The peak current (i_p) obeys the power law, as described in Eq. 1 [46]:

  (1)

  where a and b are constants, and b-value is obtained by linearly fitting log(i) and log(v). The b = 0.5 and b = 1 respectively indicates a semi-infinite diffusion process and a surface-controlled capacitive kinetics. Fig. 3e displays the fitting results, and the b-values of rGO@Fe₂O₃ (0.75/0.8) are close to 1, indicating an apparent capacitive charge storage process.

  The quantitative determination of the capacitive contribution to the total capacity was conducted according to the following Eq. 2 [47]:

  (2)

  where the current response i(V) at a specific potential could be divided into capacitive contribution (k₂v) and diffusion-controlled part (k₁v^1/2). Fig. 3f shows the proportion of capacitive contribution at different scan rates, with all the capacitive contributions greater than 50%. The capacitive contribution for the rGO@Fe₂O₃ is 75.7% at 1 mV/s (Fig. S9d in Supporting information). These results demonstrate that capacitive contributions dominate the total capacity.

  The SEM and TEM were used to investigate the morphology changes of Fe₂O₃ particles during sodiation/desodiation. Figs. 4a and b present the morphological changes of Fe₂O₃ nanocubes in the rGO@Fe₂O₃ sponge from the fresh to the fully discharged state. The metallic iron is observed at the fully discharged state (Fig. S10 in Supporting information). The nanocubes remain well dispersed on the rGO sheet without agglomeration and maintain an intact morphology. Besides, the nanocube size in the rGO@Fe₂O₃ sponge exhibits an increasing tendency (Fig. 4c). The primary size is in the range of 80–140 nm in the initial state, which changes to that in the range of 90–270 nm after deep discharge. The average size-distribution increases from the initial 112.18 nm to 186.5 nm. After desodiation, the lattice fringes of Fe₂O₃ recovered, but the lattice fringes of Fe still present (Figs. S11a-c in Supporting information). This partially irreversible transformation process affects the first Coulomb efficiency of the electrode. After desodiation, the particles would shrink, with a uniform distribution of 144.92 nm (Fig. S11d in Supporting information). The above results indicate that the sodiated Fe₂O₃ nanocubes in the rGO@Fe₂O₃ sponge maintain an intact morphology and do not pulverized. In comparison, the nanocubes in bare Fe₂O₃ are clearly deformed at fully discharged state (Figs. 4d and e). As shown in Fig. 4f, the primary size is in the range of 90–190 nm in the initial state, which changes to in the range of 30–140 nm after deep discharge. The uniform distribution decreases from the initial 142.27 nm to 79.78 nm, meaning that more small particles appear. Thus, bare Fe₂O₃ nanocubes without the protection of rGO sheets undergo a significant pulverization process. The above results indicate that the unique 3DGF effectively buffers the volume expansion of particles and inhibits their pulverization.

  Figure 4

  

  Figure 4.  Morphological changes characterization of Fe₂O₃ particles. TEM images of rGO@Fe₂O₃ sponge (a) at fresh states and (b) fully discharged states for Na-ion storage. (c) The size distribution of Fe₂O₃ nanocubes in rGO@ Fe₂O₃ sponge at fresh and fully discharged state. SEM images of pure Fe₂O₃ particles (d) at fresh states and (e) fully discharged states. (f) The size distribution of bare Fe₂O₃ particles at fresh and fully discharged states. (g) Schematic of structure changes of rGO@Fe₂O₃ and Fe₂O₃ after sodiation.

  DownLoad: Full-Size Img PowerPoint

  Based on the above discussion, the 3DGF with a continuous and hierarchical porous network enables spatial confinement to achieve outstanding results, which is visually demonstrated for the first time by the results of SEM and TEM in Fig. 4. The 3DGF not only acts as a structural framework to suppresses particle pulverization but also serves as an ionic/electronic conductive network to improve the electrochemical reaction kinetics (Fig. 4g). As a result, the rGO@Fe₂O₃ displays an excellent rate capability and cycling stability.

  It has been found that rGO@Fe₂O₃ in this work exhibits a superior specific capacity, which is favorable for obtaining full batteries with higher energy density. The coin-type Na₃V₂(PO₄)₃@C//rGO@Fe₂O₃ NIB was assembled. The anode material was pre-activated first (Fig. S12 in Supporting information). The test parameters, such as energy density, power density and specific capacity, are calculated based on the total mass of the cathode and anode materials. It delivers a specific capacity of 115 mAh/g at 0.1 A/g (Figs. S13a and b in Supporting information). At a high specific current of 4 A/g and 20 A/g, the specific capacity is 43.7 mAh/g and 30 mAh/g. As shown in Fig. S13c (Supporting information), the Ragone plots of Na₃V₂(PO₄)₃@C//rGO@Fe₂O₃ NIB exhibit a maximum energy density of 265.3 Wh/kg. A high energy density of 53.7 Wh/kg could still be obtained after 36 s of discharging. Fig. S13d (Supporting information) shows the long-term cycling performance of the full cell. After 200 cycles, the specific capacity of the full cell is 32 mAh/g.

  In summary, a self-supporting 3D graphene network decorated with Fe₂O₃ nanocubes is rationally crafted, which can be directly used as an electrode without additional binders and conductive additives. Such unique hierarchical porous architecture consisted of rGO sheets enables spatial confinement to achieve outstanding results, which can not only accommodate volume changes of sodiated Fe₂O₃ nanocubes, but also ensure favorable transport kinetics of electrons and Na⁺. Consequently, the high D_Na of rGO@Fe₂O₃ during the entire discharging/charging process is about 2.12 × 10⁻⁸/3.19 × 10⁻⁸ cm²/s, and the Na-ion storage in rGO@Fe₂O₃ sponge exhibits capacitive behavior. The rGO@Fe₂O₃ sponge anode delivers superior cycling performance (327.32 mAh/g after 2000 cycles) and superior rate capability (162 mAh/g at 20 A/g) for Na⁺ storage, which is better than the most reported results. This work provides a universal strategy for constructing a 3D conductive network to optimize the electrochemical performance of conversion-type electrodes.

  Declaration of competing interest

  The authors declare that they have no known competing finan- cial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statementt

  Jun Dong: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Senyuan Tan: Writing – review & editing, Validation, Investigation, Data curation. Sunbin Yang: Methodology, Data curation. Yalong Jiang: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Ruxing Wang: Validation, Resources, Project administration, Funding acquisition. Jian Ao: Validation. Zilun Chen: Validation. Chaohai Zhang: Validation. Qinyou An: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Xiaoxing Zhang: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

  Acknowledgments

  This work is supported by National Natural Science Foundation of China (Nos. 52307239, 52102300, 52207234), the Natural Science Foundation of Hubei Province (Nos. 2022CFB1003, 2021CFA025).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110010.

  
  
• 1. [1]

     Q. Wei, X. Chang, D. Butts, et al., Nat. Commun. 14 (2023) 7. doi: 10.1038/s41467-022-35617-3

  2. [2]

     C. Vaalma, D. Buchholz, M. Weil, et al., Nat. Rev. Mater. 3 (2018) 18013. doi: 10.1038/natrevmats.2018.13

  3. [3]

     L. Kitsu Iglesias, E.N. Antonio, T.D. Martinez, et al., Adv. Energy Mater. 13 (2023) 2302171. doi: 10.1002/aenm.202302171

  4. [4]

     T. Long, P. Chen, B. Feng, et al., Chin. Chem. Lett. 35 (2024) 109267. doi: 10.1016/j.cclet.2023.109267

  5. [5]

     Z. Li, G. Zhou, S. Li, et al., Joule 7 (2023) 2609–2621. doi: 10.1016/j.joule.2023.10.003

  6. [6]

     S. Qiao, Q. Zhou, M. Ma, et al., ACS Nano 17 (2023) 11220–11252. doi: 10.1021/acsnano.3c02892

  7. [7]

     Q. Li, H. Li, Q. Xia, et al., Nat. Mater. 20 (2021) 76–83. doi: 10.1038/s41563-020-0756-y

  8. [8]

     M. Zhang, X. Liu, J. Gu, et al., Chin. Chem. Lett. 34 (2023) 108471. doi: 10.1016/j.cclet.2023.108471

  9. [9]

     F. Kong, Y. Ge, S. Tao, et al., Chin. Chem. Lett. 35 (2023) 108552. doi: 10.1016/j.cclet.2023.108552

  10. [10]

      A. Yin, L. Yang, Z. Zhuang, et al., Energy Storage. 2 (2020) e132. doi: 10.1002/est2.132

  11. [11]

      B. Philippe, M. Valvo, F. Lindgren, et al., Chem. Mater. 26 (2014) 5028–5041. doi: 10.1021/cm5021367

  12. [12]

      N.T. Aristote, K. Zou, A. Di, et al., Chin. Chem. Lett. 33 (2022) 730–742. doi: 10.1016/j.cclet.2021.08.049

  13. [13]

      B. Papandrea, X. Xu, Y. Xu, et al., Nano Res. 9 (2016) 240–248. doi: 10.1007/s12274-016-1005-1

  14. [14]

      T. Li, A. Qin, L. Yang, et al., ACS Appl. Mater. Interfaces 9 (2017) 19900–19907. doi: 10.1021/acsami.7b04407

  15. [15]

      Y. Zhang, Q. Wang, K. Zhu, et al., Chem. Eng. J. 428 (2022) 131204. doi: 10.1016/j.cej.2021.131204

  16. [16]

      C. Zheng, X. Xu, Q. Lin, et al., Carbon N Y 176 (2021) 31–38. doi: 10.1016/j.carbon.2021.01.122

  17. [17]

      P. Ma, Z. Zhuang, J. Cao, et al., ACS Appl. Energy Mater. 5 (2022) 6417–6422. doi: 10.1021/acsaem.2c00860

  18. [18]

      Z. Zhuang, L. Yang, B. Ju, et al., Energy Storage 2 (2020) e109. doi: 10.1002/est2.109

  19. [19]

      A. Kormanyos, E. Kecsenovity, A. Honarfar, et al., Adv. Funct. Mater. 30 (2020) 2002124. doi: 10.1002/adfm.202002124

  20. [20]

      L. Shi, Y. Li, F. Zeng, et al., Chem. Eng. J. 356 (2019) 107–116. doi: 10.1016/j.cej.2018.09.018

  21. [21]

      M. Chen, D. Niu, J. Mao, et al., ACS Appl. Energy Mater. 4 (2021) 5888–5896. doi: 10.1021/acsaem.1c00691

  22. [22]

      H. Usui, Y. Domi, E. Iwama, et al., Mater. Chem. Phys. 272 (2021) 125023. doi: 10.1016/j.matchemphys.2021.125023

  23. [23]

      Q. Wang, Y. Ma, L. Liu, et al., Nanomaterials 10 (2020) 782. doi: 10.3390/nano10040782

  24. [24]

      Y. Chen, X. Yuan, C. Yang, et al., J. Alloys Compd. 777 (2019) 127–134. doi: 10.1145/3294109.3295645

  25. [25]

      T. Hou, X. Sun, D. Xie, et al., Chem. Eur. J. 24 (2018) 14786–14793. doi: 10.1002/chem.201802916

  26. [26]

      M. Fiore, G. Longoni, S. Santangelo, et al., Electrochim. Acta 269 (2018) 367–377. doi: 10.1016/j.electacta.2018.02.161

  27. [27]

      Z. Qiang, Y. Chen, B. Gurkan, et al., Carbon N Y 116 (2017) 286–293. doi: 10.1016/j.carbon.2017.01.093

  28. [28]

      G.D. Park, J.S. Cho, J.K. Lee, et al., Sci. Rep. 6 (2016) 22432. doi: 10.1038/srep22432

  29. [29]

      H. Li, L. Xu, H. Sitinamaluwa, et al., Compos. Commun. 1 (2016) 48–53. doi: 10.1038/nmat4465

  30. [30]

      D. Li, J. Zhou, X. Chen, et al., ACS Appl. Mater. Interfaces 8 (2016) 30899–30907. doi: 10.1021/acsami.6b09444

  31. [31]

      Z. Zhang, Y. Wang, S. Chou, et al., J. Power Sources 280 (2015) 107–113. doi: 10.1504/IJCNDS.2015.070290

  32. [32]

      Z. Wu, Y. Zhong, J. Liu, et al., J. Mater. Chem. A 3 (2015) 10092–10099. doi: 10.1039/C5TA01334H

  33. [33]

      Y. Fu, Q. Wei, X. Wang, et al., J. Mater. Chem. A 3 (2015) 13807–13818. doi: 10.1039/C5TA02994E

  34. [34]

      S. Wang, W. Wang, P. Zhan, et al., ChemElectroChem 1 (2014) 1636–1639. doi: 10.1002/celc.201402208

  35. [35]

      M. Valvo, F. Lindgren, U. Lafont, et al., J. Power Sources 245 (2014) 967–978. doi: 10.1016/j.jpowsour.2013.06.159

  36. [36]

      C. Li, H. Su, K. Zhang, et al., J. Alloys Compd. 858 (2021) 157714. doi: 10.1016/j.jallcom.2020.157714

  37. [37]

      Y. Chen, Z. Guo, B. Jian, et al., Nanomaterials 9 (2019) 1770. doi: 10.3390/nano9121770

  38. [38]

      S. Meng, D. Zhao, L. Wu, et al., J. Alloys Compd. 737 (2018) 130–135. doi: 10.1016/j.jallcom.2017.12.077

  39. [39]

      T. Guo, H. Liao, P. Ge, et al., Mater. Chem. Phys. 216 (2018) 58–63. doi: 10.1016/j.matchemphys.2018.05.054

  40. [40]

      M. Li, C. Ma, Q.C. Zhu, et al., Dalton Trans. 46 (2017) 5025–5032. doi: 10.1039/C7DT00540G

  41. [41]

      B. Sun, S.J. Bao, J.Le Xie, et al., RSC Adv. 4 (2014) 36815. doi: 10.1039/C4RA04686B

  42. [42]

      Z. Jian, B. Zhao, P. Liu, et al., Chem. Commun. 50 (2014) 1215–1217. doi: 10.1039/C3CC47977C

  43. [43]

      Y. Shen, Y. Jiang, Z. Yang, et al., Adv. Sci. 9 (2022) e2104504. doi: 10.1002/advs.202104504

  44. [44]

      Z. Zhuang, F. Zhang, D. Gandla, et al., ACS Appl. Mater. Interfaces 15 (2023) 38530–38539. doi: 10.1021/acsami.3c08766

  45. [45]

      Z. Zhuang, F. Zhang, Y. Zhou, et al., Materials Today Energy 30 (2022) 101192. doi: 10.1016/j.mtener.2022.101192

  46. [46]

      V. Augustyn, J. Come, M.A. Lowe, et al., Nat. Mater. 12 (2013) 518–522. doi: 10.1038/nmat3601

  47. [47]

      V. Augustyn, P. Simon and B. Dunn, Energy Environ. Sci. 7 (2014) 1597. doi: 10.1039/c3ee44164d

• 

  Figure 1  Schematic of the fabrication and structural characterization of rGO@Fe₂O₃ sponge. (a) Schematic illustration for the synthesis of rGO@Fe₂O₃ sponge. (b, c) SEM and TEM images of rGO@Fe₂O₃ sponge. (d, e) Nitrogen adsorption and desorption isotherm and pore-size distribution of rGO@Fe₂O₃ sponge.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Electrochemical performances of rGO@Fe₂O₃ sponge and Fe₂O₃. (a) The galvanostatic charge-discharge curves of rGO@Fe₂O₃ sponge at 0.1 A/g. (b, c) CV curves of rGO@Fe₂O₃ and Fe₂O₃ at 0.2 mV/s at first three cycles. (d) Long-term cycles at 1 A/g of rGO@Fe₂O₃. (e) Cycling performance comparison of current work to the state of the art. (f) Rate performance of rGO@Fe₂O₃ and Fe₂O₃ at different current densities. (g) Rate performance comparison of current work to the state of the art.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Electrochemical kinetic analysis. (a) GITT profiles of discharge and charge processes of rGO@Fe₂O₃ and Fe₂O₃ measured after three cycles at the current density of 0.02 A/g. (b, c) Na⁺ diffusion coefficients as a function of the states of charging process and Nyquist plots of rGO@ Fe₂O₃ and Fe₂O₃ at fully charged state. (d) CV curves of rGO@Fe₂O₃ at the scan rates ranging from 0.2 mV/s to 1 mV/s. (e)The normalized peak currents versus scan rate to determine the b value for anodic and cathodic peaks of rGO@Fe₂O₃. (f) The proportion of diffusion−controlled and capacitive contributions to the total charge storage at different scant rates.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Morphological changes characterization of Fe₂O₃ particles. TEM images of rGO@Fe₂O₃ sponge (a) at fresh states and (b) fully discharged states for Na-ion storage. (c) The size distribution of Fe₂O₃ nanocubes in rGO@ Fe₂O₃ sponge at fresh and fully discharged state. SEM images of pure Fe₂O₃ particles (d) at fresh states and (e) fully discharged states. (f) The size distribution of bare Fe₂O₃ particles at fresh and fully discharged states. (g) Schematic of structure changes of rGO@Fe₂O₃ and Fe₂O₃ after sodiation.

  下载: 全尺寸图片 幻灯片
•