Bamboo‐Jointed Bismuth Vanadate Structural Engineering and In Situ Atomic‐Scale Insights into Anisotropy and Sodium Ion Storage

ABSTRACT

In grid‐scale energy storage applications, sodium‐ion batteries have gained considerable attention due to the abundance of sodium resources, lower cost, and stable wide‐temperature performance. Monoclinic bismuth vanadate, composed of Bi─O octahedra and V─O tetrahedra interconnected by shared vertices to form a layered structure, facilitates the reversible insertion and extraction of sodium ions. However, owing to the lack of direct atomic‐level observation, the precise sodium storage mechanism remains unclear. Here, we utilized electrospinning to fabricate bamboo‐jointed BiVO₄ nanorods, whose segmented features effectively mitigate axial stress transfer, relieve volumetric strain, and suppress pulverization. Furthermore, we employ in situ transmission electron microscopy (TEM), combined with density functional theory (DFT) calculations, to investigate the structural evolution of bamboo‐jointed BiVO₄ during (de)sodiation. The anisotropic expansion of bamboo‐jointed BiVO₄ during initial sodium insertion is revealed for the first time, whereas its morphology becomes more isotropic in subsequent cycles. This transformation is attributed to the low stability of the alloyed Na‐Bi product, which promotes particle fusion. Additionally, the detailed electrochemical performance and cycling mechanism are also comprehensively elucidated. This work not only bridges the gap in understanding the sodium storage mechanism of BiVO₄ but also offers valuable insights for designing tailored morphologies for high‐performance sodium‐ion storage.

This work reveals anisotropic Na‐ion insertion and structure evolution in bamboo‐jointed BiVO₄ nanorods through in situ TEM and DFT calculation. The segmented architecture mitigates stress, suppresses pulverization, and transforms toward isotropic behavior upon cycling, providing new mechanistic insights for designing high‐performance sodium‐ion storage materials.

1. Introduction

Lithium‐ion batteries (LIBs), with their high energy density, long cycle life, and mature industry chain, are the mainstream energy storage technology for consumer electronics and electric vehicles [1, 2, 3]. However, for grid‐scale energy storage, LIBs face several challenges, including limited lithium resources and price volatility, an underdeveloped recycling system, and safety concerns, which restrict their applicability in long‐duration, low‐cost, large‐scale storage [4, 5]. Sodium‐ion batteries (SIBs), benefiting from abundant sodium resources, superior cost potential, and inherent advantages in safety and low‐temperature characteristics, have recently emerged as a promising supplement and alternative for large‐scale energy storage [6, 7, 8, 9, 10]. Nevertheless, due to the intrinsic differences in atomic properties between sodium and lithium (e.g., larger ionic radius and higher migration mass), sodium ions encounter difficulties during insertion/extraction in some electrodes, often leading to hindered diffusion thermodynamics and slower reaction kinetics [11, 12, 13, 14]. Consequently, discovering suitable sodium‐ion electrodes that offer both high reversible capacity and strong structural stability remains a significant challenge.

Among various electrodes, alloying anodes (such as Si, Sn, Sb, P, V, Ti, and Bi) and their oxides/sulfides have attracted considerable attention for sodium‐ion batteries compared to intercalation and pure conversion‐type systems, as their multi‐electron reactions deliver higher capacities [15, 16, 17, 18, 19, 20]. For instance, Sb alloying to Na₃Sb provides a theoretical capacity of about 660 mAh g⁻¹, whereas hard carbon typically exhibits around 200–300 mAh g⁻¹ [21]. Moreover, alloying anodes have a more suitable low potential plateau (around 0.3–0.8 V vs. Na/Na⁺), which increases the full‐cell operating voltage and mitigates electrolyte oxidation stress. The final alloying product of Sn, Na₁₅Sn₄, demonstrates excellent sodium‐ion storage performance with yolk–shell/carbon skeleton buffering structure [22]. Derivatives such as SnO₂, SnS, and Bi₂O₃ also exhibit high sodium‐storage capability through the conversion‐alloying mechanism. However, due to the relatively large size of sodium ions, the sodiation process in these electrodes often results in significant volume expansion, leading to pulverization, loss of electrical contact, and capacity degradation. In existing research, substantial effort relies on construction strategies such as complex hierarchical pore structures, intricate hollow/core‐shell structures, and complex micro/nano frameworks. While these approaches have indeed improved the structural stability and electrochemical performance of electrodes to a certain extent, they often involve multi‐step chemical reactions, tedious templating/etching processes, and strict condition controls. Consequently, there remains a significant gap between these methods and the requirements for large‐scale, low‐cost, and sustainable manufacturing.

Monoclinic bismuth vanadate consists of Bi─O octahedra and V─O tetrahedra connected through shared vertices to form a layered structure, facilitating ion insertion and extraction [23, 24, 25, 26, 27]. In 1989, Pasquali and Pistoia first used BiVO₄ as a cathode for lithium batteries [28]. In 2017, Sottmann et al. reported BiVO₄ as a stable alloying‐type anode for SIBs, exhibiting a high specific capacity (367 mAh g⁻¹) and superior rate performance [29]. Compared with hard carbon, which typically delivers only 200–300 mAh g⁻¹ [21], and metallic Bi, which suffers from severe volume expansion despite its high theoretical capacity, BiVO₄ is particularly attractive because it offers a more balanced combination of capacity, working potential, and structural stability through its conversion–alloying reaction pathway [30, 31, 32, 33]. However, effective strategies for mitigating volume changes caused by sodium insertion remain insufficient. In particular, the pronounced irreversible process during the first cycle and the formation/disappearance of the Bi/Na_xBi phases in subsequent cycles leave significant uncertainties regarding the atomic‐scale reaction pathways and interfacial evolution. Especially, the uncertainties surrounding atomic‐scale sodium‐ion migration behavior and intermediate phases obscure the bond breaking and formation pathways of BiVO₄ during the initial discharge, as well as the precise depolymerization sequence of the V─O substructure.

In this work, we fabricate bamboo‐jointed BiVO₄ anodes using a simple electrospinning technique. This unique bamboo‐jointed architecture effectively buffers the volume stress generated during sodiation. Moreover, by integrating in situ TEM and electrochemical platforms, we constructed all‐solid‐state nanoscale SIBs, enabling real‐time visualization of BiVO₄ during (de)sodiation and cycling. Combined with DFT calculations, we revealed for the first time the anisotropic expansion behavior of bamboo‐jointed BiVO₄ anode during initial sodiation. The DFT calculations indicate that the low stability of the Na–Bi alloyed product facilitates the initial fusion and reconstruction of the reaction products during sodiation, which helps explain the reduced structural anisotropy and improved stability in subsequent cycles. In addition, stepwise phase transitions during sodiation were identified through HRTEM imaging and time‐resolved Fast Fourier Transform (FFT). The detailed desodiation behavior, cycling mechanism and electrochemical performance are also clearly elucidated. This work not only systematically reveals the structural evolution and electrochemical reaction pathways of BiVO₄ during sodium‐ion insertion/extraction, bridging the gap in sodium storage mechanism, but also provides valuable insights for the specific morphological design and optimization of high‐performance sodium‐ion storage.

2. Results and Discussion

We introduce a bamboo‐jointed BiVO₄ anode fabricated by electrospinning to mitigate sodiation‐induced stress, as shown in Figure 1a. The segmented nanorods feature widths of 200–300 nm, and their lengths can be controlled by electrospinning, ranging from several hundred nanometers to several micrometers (Figure 1b). This architecture, analogous to previously reported SiOx/C and CNTs@Sn₄P₃@C designs, may mitigate axial stress propagation, accommodate volume expansion during cycling, and suppress pulverization, thereby preserving structural integrity and electrical connectivity [34, 35, 36]. The XRD result in Figure S1a indicate that the main diffraction peaks of the prepared sample perfectly match the crystal planes of monoclinic BiVO₄ (space group I2/b(15), PDF#00‐014‐0688), demonstrating its good crystallinity. The corresponding XPS characterization is shown in Figure S1b. Two strong peaks at binding energies of 157.2 and 162.4 eV correspond to Bi4f _7/2 and Bi4f _5/2, respectively, confirming the presence of Bi³⁺. Furthermore, two strong peaks corresponding to V2 _p3/2 and V2 _p1/2 indicate that V⁵⁺ in BiVO₄. The morphology and structure of BiVO₄ nanorods were further characterized by TEM. Figure 1c shows the HRTEM image of BiVO₄ nanorods, where clear lattice fringes are visible with a lattice spacing of 2.24 Å, corresponding to the (112) plane of BiVO₄. The SAED patterns of BiVO₄ in Figure 1d and Figure S2 exhibit regularly arranged diffraction spots, indicating high crystallinity and single‐crystal‐like characteristics in the selected region, which is consistent with the XRD results. However, considering the segmented bamboo‐jointed morphology of the electrospun nanorods, these data are insufficient to support an unambiguous assignment of continuous single crystallinity and the exact crystallographic orientation along the entire nanorod length. The elemental distribution of the pristine BiVO₄ nanorods is shown in Figure 1e, indicating that the Bi, V, and O elements are uniformly distributed throughout the sample. Figure 1f is a schematic diagram of the atomic structure of monoclinic BiVO₄, showing that BiVO₄ consists of VO₄ tetrahedra and BiO₈ octahedra, in which similar metal atoms are interconnected in a continuous zigzag configuration. Additionally, because Bi and V atoms alternate along the axis, the anode exhibits a layered structural feature that facilitates sodium‐ion transport. An all‐solid‐state nanoscale SIBs assembled in the TEM is shown in Figure 1g. The corresponding experimental setup is depicted in Figure 1h. Bamboo‐jointed BiVO₄ nanorods were used as the working electrode, while a tungsten tip with freshly deposited sodium metal functioned as the counter electrode.

FIGURE 1.

Morphology and microstructure characterization of pristine BiVO₄ nanorods. (a) Electrospinning technique for preparing bamboo‐jointed BiVO₄ nanorods, (b) SEM of bamboo‐jointed BiVO₄ nanorods, (c) HRTEM image, (d) SAED pattern, (e) EDS elemental mappings, (f) Atomic crystal structure models of layered monoclinic BiVO₄, (g) Schematic illustration of the all‐solid‐state nanoscale sodium‐ion battery, (h) The corresponding experimental setup.

Figure 2a₁ shows a TEM image of the pristine BiVO₄ anode. A potential of −3 V was applied to the BiVO₄ anode to initiate sodiation, resulting in a series of sodiation images, as shown in Figure 2a₁‐a₆ and Movie S1. Sodium ions diffused from the electrode contact point into the interior of the BiVO₄ anode, with the green arrows indicating the diffusion direction of the sodium ions. After sodium ion intercalation, there was a contrast difference between the sodiated and unsodiated regions of the BiVO₄. This boundary is termed the reaction front (RF), as indicated by the green dashed line. The progression of the RF allowed real‐time observation of the sodiation reaction. As sodium ions continued to intercalate, the RF took on a “U” shape, attributed to the faster surface diffusion of sodium ions compared with that in the interior. With further sodiation, we observed the expansion and bending of BiVO₄ segmented nanoparticles ②. Upon complete sodiation, the overlapping area at the junction of the two particles increased, and the change in the angle at the junction dispersed the stress generated by the volume expansion of the anode along the original direction. In addition, volume changes during the sodiation were also observed. Because TEM images are limited in capturing the three‐dimensional volume variations, we quantified the relationship between the projected reaction region area and reaction time, as shown in Figure 2b. Similar to LIBs, the image exhibits a “square root function” trend. It can be observed that as sodium ions intercalate, the sodiated region gradually increases in a linear growth pattern, but after 300 s, the rate of increase slows down. Furthermore, the bamboo‐jointed BiVO₄ maintained its structural integrity throughout the entire sodiation without any fragmentation. This stability can be attributed to the bamboo‐jointed architecture designed by the electrospinning process, whose segmented characteristics significantly reduce axial stress transfer and release the volume stress induced by sodiation.

FIGURE 2.

The sodiation process of the BiVO₄ anode. (a₁‐a₆) Time‐lapse TEM images, (b) Area analysis of the sodiated region of BiVO₄ nanorod in (a) as a function of time, (c₁, c₂) TEM images of another BiVO₄ nanorod before and after sodiation, (d) Transverse and longitudinal dimension elongation of sodiated BiVO₄ segments, (e–g) DFT calculations were performed to investigate the intercalation behavior of Na ions in BiVO₄, The resulting stable atomic structures (Na₂BiVO₄, Na₄BiVO₄, and Na₆BiVO₄) were observed along the [100] direction, (h) HAADF‐STEM image and EDS elemental mappings of the sodiated BiVO₄ segment.

To comprehensively analyze the sodium‐ion transport behavior of BiVO₄, we examined its overall bamboo‐jointed structure of BiVO₄ before and after sodiation. Figure 2c₁ shows the pristine BiVO₄, while Figure 2c₂ presents its morphology after sodium intercalation. We statistically analyzed the width and length of three particles in the segmented structure before and after sodiation, and calculated the transverse and longitudinal expansion rates, as shown in Figure 2d. Due to the inevitable bending of the BiVO₄ electrode during sodiation, the size statistics on the side closer to the sodium source are less accurate. Nevertheless, it can still be observed that the transverse expansion rates of the three particles from the near‐sodium end to the far‐sodium end remain around 9%. In contrast, the longitudinal expansion rate differs from the transverse one, indicating a pronounced anisotropic expansion of the BiVO₄ anode. To confirm that this phenomenon is not an isolated case, the structural evolution of another BiVO₄ nanorod was investigated during sodium insertion. The time‐series images of sodiation are shown in Figure S3a₁–a₈ . After sodium insertion, sodium ions migrate along the direction indicated by the green dashed‐arrow. As the reaction proceeds, the sodiated regions become darker in contrast, and a distinct RF (marked by a green arrow) is observed. Sodium insertion causes lattice changes accompanied by volume expansion. After full sodiation, the edges of the BiVO₄ particles become rounded, and the boundaries between particles gradually disappear. However, the BiVO₄ overall does not exhibit common fragmentation features such as holes or cracks. These findings align with the previously described sodiation behavior. The corresponding changes in the sodiated area over time in Figure S4a are also consistent with the results in Figure 2b. In addition, the transverse and longitudinal expansion ratios of particles ① and ② after sodiation were calculated and are shown in Figure S4b. The results reveal that the longitudinal expansion rate is significantly higher than the transverse one, further validating the anisotropic behavior of BiVO₄ during sodiation. Moreover, DFT calculations were conducted to further elucidate the local atomic structures associated with sodiation degrees. The corresponding initial structural models are presented in Figure 1f and the lattice parameters of BiVO₄ are a = 5.15 Å, b = 11.71 Å, c = 5.15 Å. Figure 2e–g display the stable atomic configurations obtained along the [100] direction (Na₂BiVO₄, Na₄BiVO₄, and Na₆BiVO₄). When a small amount of sodium ions is inserted the BiVO₄ formula unit, both lattice parameters a and c exhibit slight contraction due to electrostatic repulsion between the inserted Na⁺ ions and the layered structure. However, this subtle change was not experimentally detected because it is transient. With the further insertion of 2 Na⁺ ions (Figure 2e), the transverse direction expands by approximately 0.68%, while the longitudinal direction exhibits a remarkable elongation of about 9.08%, as shown in Figure S5. This result is consistent with the anisotropic expansion behavior observed experimentally. Upon further insertion of 4 Na⁺ ions into BiVO₄, the longitudinal expansion rate reaches 13.97%, whereas the transverse direction displays a moderate expansion of 2.51%. Furthermore, with the insertion of 6 Na⁺ ions, the longitudinal expansion rate remains nearly constant, while the transverse expansion increases to 6.08%. These DFT results clearly demonstrate asymmetric expansion within the a–b plane of BiVO₄ at the single‐unit‐cell scale and provide a theoretical explanation for the in situ TEM observations described above. At this stage, a direct quantitative comparison of lattice parameters and Na⁺ content between experiment and DFT remains difficult, because the intermediate Na_xBiVO₄ phase is highly transient and rapidly loses its crystalline features during in situ sodiation. Under these conditions, we believe that the anisotropic expansion trend provides the most meaningful basis for comparison, and in this respect the experimental observations are consistent with the DFT results.

In addition, EDS elemental analysis was performed on the sodiated BiVO₄ anode, selecting the area shown by the purple box in Figure 2c₂ . The elements Bi, V, O, and Na are uniformly distributed, indicating that the BiVO₄ anode is fully sodiated, as illustrated in Figure 2h. Although the exact crystal orientation along the nanorod length cannot be determined here, both experiment and DFT support the intrinsic anisotropic sodiation behavior of BiVO₄.

To deeply investigate the structural evolution mechanism of BiVO₄ during sodiation, we employed in situ HRTEM to observe the phase transitions occurring throughout the reaction. Figure 3 and Movie S2 show HRTEM images at various sodiation stages, along with enlarged views of local regions and the corresponding Fast Fourier Transform (FFT) patterns. As shown in Figure 3a, the pristine BiVO₄ anode exhibits clear lattice fringes in the analyzed local region, indicating a highly crystalline structure before sodiation, with lattice fringes of 2.55 Å corresponding to the (002) plane. Upon sodium ions insertion, the lattice spacing of the (020) plane gradually expands, indicating that the intercalation of sodium ions causes a slight expansion of the host lattice. The magnified HRTEM image in Figure 3b reveals that BiVO₄ starts to transform into an amorphous structure. Its corresponding FFT pattern exhibits amorphous rings, indicating that the insertion of sodium ions disrupts the original crystal structure of BiVO₄ and generates a disordered Na _x BiVO₄ phase. As more sodium ions are introduced, the diffuse diffraction rings gradually fade, and new diffraction spots appear. These new spots correspond to emerging phases of Bi, Na₂Bi, and Na₃VO₄, suggesting that both conversion and alloying reactions occurred at this stage, as shown in Figure 3c. The corresponding FFT in Figure 3d shows that in the late stage of sodiation, Bi nanoparticles continue to react with sodium ions to form Na₃Bi alloy. With the insertion of excess sodium, the overall structural differences before and after sodiation are clearly displayed in Figure 3a,d. The HRTEM images demonstrate that the abundant nanoparticles generated by the Na₃Bi alloying reaction reflect the structural evolution and the formation of new phases during sodiation.

FIGURE 3.

Structural evolution of the BiVO₄ anode at different sodiation stages. (a−d) Time‐lapse structural evolution taken 0 (a), 104 (b), 320 (c), and 706 s (d), showing representative TEM images (top), HRTEM images (middle), and FFT patterns (bottom).

Through HRTEM images and FFT analysis, the transition of the BiVO₄ crystal structure from order to disorder, along with the formation of a new phase after sodiation, was revealed. Initially, driven by negative pressure, sodium ions are inserted into BiVO₄, leading to an intercalation reaction that forms the amorphous phase Na _x BiVO₄. The sodiation reaction at this stage is:

$$\mathrm{BiV}{\mathrm{O}}_4+x\mathrm{N}{\mathrm{a}}^{+}+x{\mathrm{e}}^{-}\to \mathrm{N}{\mathrm{a}}_{\mathrm{x}}\mathrm{BiV}{\mathrm{O}}_4$$ (1)

Further intercalation of sodium ion converts Na _x BiVO₄ into Bi and a sodium vanadate phase, which is assigned as Na₃VO₄ based on the combined in situ TEM and in situ XRD results. The corresponding reaction equation is:

$$\mathrm{N}{\mathrm{a}}_{\mathrm{x}}\mathrm{BiV}{\mathrm{O}}_4+\left(3-\mathrm{x}\right)\mathrm{N}{\mathrm{a}}^{+}+\left(3-\mathrm{x}\right){\mathrm{e}}^{-}\to \mathrm{Bi}+\mathrm{N}{\mathrm{a}}_3\mathrm{V}{\mathrm{O}}_4$$ (2)

Subsequently, the sodiation reaction proceeds as sodium ions alloy with Bi to produce a NaBi alloy. The corresponding reaction equations for this stage are presented below:

$$\mathrm{Bi}+2\mathrm{N}{\mathrm{a}}^{+}+2{\mathrm{e}}^{-}\to \mathrm{N}{\mathrm{a}}_2\mathrm{Bi}$$ (3)

$$\mathrm{N}{\mathrm{a}}_3\mathrm{Bi}+\mathrm{N}{\mathrm{a}}^{+}+{\mathrm{e}}^{-}\to \mathrm{N}{\mathrm{a}}_3\mathrm{Bi}$$ (4)

After sodium ion intercalation, a series of sodiation reactions occur, resulting in the final sodiated products of Na₃Bi and a sodium vanadate phase reasonably assigned as Na₃VO₄. Another set of HRTEM sodiation reactions is presented in Figure S6, and the observed results are consistent with those shown in Figure 3. The in situ XRD results (Figure S7) further support the phase‐evolution sequence inferred from the in situ TEM analysis. However, due to the highly transient and poorly crystalline of the intermediate Na_xBiVO₄ phase, its exact lattice parameters cannot be reliably extracted for direct comparison with the DFT models.

Therefore, a schematic diagram of the microstructure evolution of BiVO₄ during sodiation is shown in Figure 4a. In the sodiation studies of the three samples, no pulverization was observed, suggesting its great potential as an anode for SIBs. The charge density difference and binding energy of the BiVO₄ anode during sodium insertion were calculated using DFT, as shown in Figure 4b.

FIGURE 4.

(a) Schematic illustration of the atomic structure and reaction mechanism during the sodiation process, (b) Charge density difference projection, (c) Cohesive energy.

Analysis of the charge density difference plot reveals that variations in color correspond to differences in charge density, where a larger color contrast indicates a stronger interphase binding force. The calculation results show that the Na─Bi alloy exhibits a small charge density difference, suggesting a weak binding interaction, while the Na₃Bi shows comparatively low structural stability. Combined with literature on the fluidity of Bi‐based alloys [37], it can be inferred that during electrochemical reactions, the instability of the Na─Bi alloy can cause the blurred boundaries of BiVO₄ nanoparticles and their tendency to integrate. More importantly, the continued morphological rearrangement observed after desodiation is more reasonably attributed to the formation of ultra‐small Bi nanoparticles. Under the in situ TEM condition, the pronounced size effect of Bi at the nanoscale, together with electron‐beam‐induced local heating and interfacial instability, may promote quasi‐liquid‐like rearrangement and coalescence of these nano‐Bi domains, leading to the observed blurred particle boundaries and integration tendency. This finding provides strong theoretical support for understanding the microscopic mechanism of morphological evolution during sodium‐ion insertion and deepens the comprehension of phase transitions and particle coalescence behavior in electrochemical reactions.

After the sodiation reaction was completed, a positive voltage of 3 V was applied to BiVO₄ to initiate desodiation. Figure S8a₁–a₆ illustrates the morphological evolution of the sodiated BiVO₄ anode during this process. Owing to the bamboo‐jointed shape of the BiVO₄ anode, the change in projected area was measured from the sodium source contact end to the position marked by the yellow dashed line to quantitatively assess the size variation during desodiation. Based on the measurement data, the area at the beginning of desodiation was 6.1 × 10⁵ nm², and after 199 s of continuous desodiation, it decreased to 5.7 × 10⁵ nm², corresponding to a shrinkage of approximately 6.48%. Figure S8b shows the trend in area change within the marked region during this process. Although the area change was relatively pronounced in the initial stage, the rate of shrinkage gradually slowed as desodiation continued. Comparing the morphologies in Figures S8a₁ and S8a₂ indicates that, although the BiVO₄ anode size changed during desodiation, sodium ions were not completely extracted. This observation suggests a possible sodium‐ion retention effect during extraction, preventing the BiVO₄ anode from fully returning to its original state.

To investigate the conversion‐alloying mechanism of the BiVO₄ anode during charge–discharge cycles, we further examined its structural evolution over the first four sodiation/desodiation processes. Figure 5a illustrates the morphological changes of the BiVO₄ anode. Observations show that the anode undergoes significant volume expansion after the first sodiation, with lateral and longitudinal dimension increases of 27% and 80%, respectively, demonstrating clear anisotropy. However, in subsequent sodiation and desodiation cycles, the volume variations are minimal, suggesting that the product formed after the first cycle enables stable and repeated sodium‐ion insertion and extraction. This occurs because, following the first sodiation, the anode is no longer the initial layered BiVO₄. In addition, the small sodium‐ion concentration gradient within the anode, along with the instability of the Na–Bi alloy that encourages coalescence of the reaction products, reduces structural anisotropy. Consequently, the anode experiences nearly isotropic dimensional changes. Figure 5b displays the area change of the false‐color‐marked region during charge and discharge. After the first electrochemical cycle, both the lateral and longitudinal dimensions of BiVO₄ remain nearly unchanged, with an area variation within 5%. Importantly, no structural fractures are observed in the BiVO₄ anode over multiple electrochemical cycles. These results demonstrate that the BiVO₄ anode maintains excellent structural stability during sodium‐ion insertion and extraction, confirming its suitability as an anode material for SIBs.

FIGURE 5.

Morphology and area evolution of the BiVO₄ anode during the four sodiation/desodiation cycles. (a) Time‐resolved TEM images showing morphological changes, (b) Variation of the BiVO₄ anode area as a function of the sodiation/desodiation cycle time.

To understand the application of the BiVO₄ anode in SIBs, we systematically investigated its macroscopic electrochemical performance. Figure 6a presents the charge–discharge curves of the BiVO₄ anode at a current density of 20 mA g^{− 1}. The initial sodiation and desodiation capacities are 924.9 and 680.5 mAh g^{− 1}, respectively, with a first‐cycle coulombic efficiency of approximately 73.6%. This initial capacity loss can be attributed to the formation of the SEI layer and the intercalation reaction of sodium ions. Figure 6b shows the cyclic voltammetry curves for the first three cycles at a scan rate of 0.3 mV s^{− 1} within a potential window of 0.0–3.0 V (vs. Na⁺/Na). The peak observed around 0.87 V during the first sodiation may correspond to the BiVO₄ conversion reaction. Additionally, the sodiation peak below 0.22 V is likely associated with electrolyte decomposition and the polarization of sodium ions on the electrode. In subsequent cycles, the initial discharge peak at 0.87 V disappears, and the remaining peaks shift toward lower potentials, suggesting that the initial conversion reaction is irreversible. As cycling progresses, the peaks in the second and third cycles remain within the same potential range, but the current intensity decreases, indicating that the alloying reaction tends to stabilize. To further evaluate the charge–discharge performance of the BiVO₄ anode under different current densities, Figure 6c presents its rate performance. Five‐cycle tests were conducted at 20, 50, 100, 200, and 500 mA g^{− 1}, yielding corresponding discharge capacities of 484.1, 413.3, 345.7, 294.7, and 296.9 mAh g^{− 1}, respectively. Although the discharge capacity decreases at higher current densities, BiVO₄ still exhibits relatively stable capacity at high current densities. After multiple cycles, the discharge capacity of BiVO₄ can still be maintained at approximately 492.4 mAh g⁻¹ at 20 mA g⁻¹, demonstrating strong cycling stability. Figure 6d presents the coulombic efficiency and capacity retention of the BiVO₄ anode at 1000 mA g⁻¹. It is evident that the capacity retention in the later cycles remains stable at around 67% throughout the test, while the coulombic efficiency consistently stays above 98%. This demonstrates that the BiVO₄ anode exhibits good electrochemical stability during long‐term cycling, consistent with the stable cycling morphology shown in Figure 5. Therefore, although some irreversible reactions occur during the first charge–discharge cycle, the performance of the BiVO₄ anode gradually stabilizes with further cycling, showing strong capacity retention and excellent electrochemical performance.

FIGURE 6.

Electrochemical performance of BiVO₄ nanorods as anode material in SIBs. (a) Galvanostatic charge–discharge curves, (b) Cyclic voltammetry profiles, (c) Electrochemical rate performance at different current densities, (d) Cycling capacity and coulombic efficiency.

3. Conclusion

In summary, we employed in situ TEM to thoroughly investigate the sodiation/desodiation process of bamboo‐jointed BiVO₄ anodes. By combining in situ HRTEM with DFT calculations, we elucidated the sodium‐ion transport behavior and the multi‐step intercalation–conversion–alloying reaction mechanism. During the initial sodiation, the BiVO₄ anode first undergoes an intercalation reaction forming an amorphous phase, followed by a conversion reaction producing Bi and Na₃VO₄, and then a continuous alloying reaction of Bi. The overall sodiation pathway is BiVO₄ → Na _x BiVO₄→ Bi + Na₃VO₄→ Na₃Bi + Na₃VO₄. Owing to the layered structure of BiVO₄, the diffusion coefficient and local ion concentration vary along different directions, leading to pronounced anisotropic characteristics in the transverse and longitudinal dimensions. The longitudinal expansion is significantly greater than the transverse one. However, in subsequent cycles, dimensional variations in different directions tend to become isotropic. During later electrochemical cycles, the electrode area change remains below 5%, and no evident defects such as fragmentation, fracture, or voids are observed, indicating excellent cycling stability. This study offers valuable insights for further exploration of the sodium‐storage mechanism in BiVO₄ anodes and for optimizing the structural stability of SIB electrodes.

4. Experimental Section

4.1. Materials Preparation and Characterization

The bamboo‐jointed BiVO₄ nanorods were fabricated by electrospinning [38]. Bi(NO₃)₃·5H₂O was dissolved in a mixed solution containing N, N‐dimethylformamide (DMF) and acetic acid, and stirred until completely clear. Subsequently, polyvinylpyrrolidone (PVP), C₁₀H₄O₅V, and ethanol were added, and stirring continued until the solution turned transparent, yielding the precursor solution. Next, the precursor solution was drawn into a syringe, connected to a soft tube needle. After electrospinning for 2 h, a precursor film was obtained. The precursor film was then placed in an oven at 150°C for pre‐oxidation for 12 h. After pre‐oxidation, the film was placed in a porcelain boat and annealed at 450°C, 500°C, and 550°C, respectively. Annealing removes the PVP, yielding a light yellow sample. The basic phase structure and surface morphology were characterized using a powder X‐ray diffractometer (XRD, Bruker D8 diffractometer) with a 2𝜃 scanning range from 20 to 70°, and a scanning electron microscopy (SEM, JSM‐7600F, JEOL, Japan) operated at 15 kV. The chemical composition and electronic states were measured using X‐ray photoelectron spectroscopy (XPS) with an Al Kα X‐ray source. Further structural, morphological analysis and elemental composition were performed using an FEI Talos F200X TEM with an X‐FEG electron source module operated at 200 keV.

4.2. In Situ Electrochemical Sodiation/Desodiation Experiments

To investigate the sodium storage mechanism in real‐time, we constructed a nanoscale solid‐state SIBs inside a TEM using an in situ electrochemical sample holder (PicoFemto, Talos F200X). A half‐copper grid loaded with bamboo‐jointed BiVO₄ nanorods was mounted on the holder and used as the working electrode in the nanoscale SIBs. A tungsten tip of suitable length was used to scrape the surface of freshly cleaned sodium metal, ensuring that the tungsten tip surface contained a trace amount of sodium source, which served as the counter electrode. The assembled electrochemical sample holder was quickly transferred to the TEM chamber. At this point, an oxide layer inevitably forms on the surface of the sodium source, acting as the solid‐state electrolyte. A voltage relative to the Na/NaO _x electrode was applied to the bamboo‐jointed BiVO₄ nanorods to drive the insertion or extraction of sodium ions, achieving sodiation and desodiation of the nanorods.

4.3. DFT Calculations

In this work, we employ DFT to calculate the ionic transport behavior, cohesive energy of each phase, and charge density difference during sodiation, thereby comparatively analyzing the structural stability of each phase and elucidating the underlying mechanisms of microstructural changes. The Vienna Ab‐initio Simulation Package (VASP) was used to perform the DFT calculations [39]. Furthermore, we adopted the generalized gradient approximation (GGA) method and the Perdew‐Burke‐Ernzerhof (PBE) functional to treat the exchange‐correlation effects [40, 41]. The cutoff energy was set to 500 eV, and the energy convergence criterion was set to 1.0 × 10⁻⁵ eV. To evaluate the structural stability, we calculated the cohesive energy of the electrochemical reaction products, as shown in the Equation (5):

$${\mathrm{E}}_{\mathrm{coh}}=\frac{{\mathrm{E}}_{{\mathrm{A}}_{\mathrm{n}}{\mathrm{B}}_{\mathrm{m}}}-\mathrm{n}{\mathrm{E}}_{\mathrm{A}}-\mathrm{m}{\mathrm{E}}_{\mathrm{B}}}{\mathrm{n}+\mathrm{m}}$$ (5)

E_coh represents the total energy of the compound. E_A is the energy of a single A atom, E_B represents the energy of a single B atom. The charge density difference in a system can be calculated using the following Equation (6).

$$\mathrm{\Delta }\rho ={\rho }_{\mathrm{total}}-\sum {\rho }_{\mathrm{i}}$$ (6)

Here, ρ_total represents the charge density of the entire system, and ρ_i represents the charge density of the i‐th atom in the system.

4.4. Electrochemical Measurements

The electrochemical performance of the bamboo‐jointed BiVO₄ anode for sodium‐ion batteries was investigated using coin cells. The working electrode slurry was prepared by dissolving the active BiVO₄, sodium alginate, and carbon black in water at a mass ratio of 7:1:2. The slurry was uniformly coated on a copper foil current collector using a glass rod, with a coating thickness of approximately 10 µm. The coated electrodes were placed on a hot plate at 80°C and dried for 12 h. After drying, the electrode sheets were cut into circular electrodes with a diameter of 10 mm and pressed under a pressure of 10 MPa. SIBs were assembled using metallic sodium as the counter electrode, the prepared BiVO₄ electrode as the working electrode, and 1 M NaPF₆ (PC with 5% FEC) solution as the electrolyte. The battery electrochemical performance tests were carried out in an argon‐filled glove box, where the oxygen and water content were below 0.01 ppm. Constant current discharge/charge cycling was performed at room temperature using a LAND CT2001A battery testing system (Wuhan, China) with a voltage range of 0.01−3.0 V (vs Na⁺/Na). Cyclic voltammetry (CV) measurements were performed at different scan rates on an Autolab potentiostat (Autolab Instruments, Netherlands).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: smll73840‐sup‐0001‐SuppMat.docx.

Supporting File 2: smll73840‐sup‐0002‐MovieS1.mp4.

Supporting File 3: smll73840‐sup‐0003‐MovieS2.mp4.

Data Availability Statement

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