Unlocking the Zn storage performance of ammonium vanadate nanoflowers as high-capacity cathodes for aqueous zinc-ion batteries via potassium ion and ethylene glycol co-intercalation engineering

Abstract

Ammonium vanadate (NH₄V₄O₁₀) is a highly promising cathode for aqueous zinc ion batteries (AZIBs) due to its tunable layered structure and high specific capacity; however, limited active sites, poor kinetics and irreversible structural collapse during cycling suppress its wide application. Herein, the engineering of the co-intercalation of K⁺ and ethylene glycol molecules is proposed to unlock the Zn storage performance of NH₄V₄O₁₀. It is found that the co-intercalation of K⁺ and ethylene glycol enlarges the interlayer spacing to 11.5 Å, induces a high level of oxygen vacancies and enhances the strength of ionic bonding in NH₄V₄O₁₀, ensuring large Zn²⁺ diffusion channels, efficient redox reactivity and strong layered-structure stability. Meanwhile, K⁺ ions and ethylene glycol also weaken the crystallinity of NH₄V₄O₁₀ and induce the transformation of microscopic morphology from strips to nanoflowers self-assembled from ultrathin nanosheets, promoting the transfer of electrons and ions and complete penetration of the electrolyte during electrochemical reactions. In addition, the band gap is significantly reduced by 0.2 eV after the co-intercalation of K⁺ and ethylene glycol, improving the electronic conductivity and decreasing the diffusion barrier of Zn²⁺. As expected, K⁺ ions and ethylene glycol co-intercalated NH₄V₄O₁₀ exhibits excellent Zn storage properties, delivering an ultrahigh capacity of 614.1 mA h g⁻¹ at 0.5 A g⁻¹ and presenting an outstanding rate performance of 472.9 mA h g⁻¹ and a high retention of 89% after 2000 cycles at 10 A g⁻¹. This work provides a useful reference for unlocking the Zn storage performance of layered V-based cathodes for AZIBs by synergistically modulating the interlayer spacing, oxygen defects and microscopic morphology of the materials.

---

1. Introduction

In recent years, rechargeable secondary batteries with high energy density and reliability have been recognized as one of the most promising technologies for energy storage and conversion.^1–5 In particular, aqueous zinc-ion batteries (AZIBs) have been considered a promising alternative to large-scale lithium-ion batteries for energy storage because of their tremendous cost competitiveness, environmental friendliness, relatively low redox potential (−0.76 V vs. the standard hydrogen electrode) and high theoretical specific capacity (820 mA h g⁻¹ or 5851 mA h cm⁻³).^6–11 However, designing cathode materials with a good balance between high capacity, excellent rate performance and good structural stability remains the main challenge in commercializing AZIBs.^6,7,12–16

Among various cathodes for AZIBs, layered vanadium-based compounds stand out due to their high theoretical capacity, multi-electron redox reactions of vanadium elements, and tunable interlayer spacing.^17–20 In general, the interlayer spacing of layered vanadium-based compounds can be reasonably modulated by introducing cations,²¹ molecules,¹⁷ or polymers²² between the layers to obtain larger ion diffusion channels and improve kinetic behavior. Among them, NH₄⁺ has a large radius and low mass, providing high mass and volume-specific capacity when it is pre-inserted into the layered vanadium-based compounds to form NH₄V₄O₁₀.²³ In addition, NH₄V₄O₁₀ consists of V₄O₁₀ units, forming an adjustable stable bilayer structure by sharing vanadium octahedral edges.^24–26 However, Zn²⁺ ions readily bind to strongly electronegative oxygen atoms, thus generating large stresses in the main structural framework during frequent insertion and detachment processes, leading to the irreversible collapse of the lamellar structure and thus greatly degrading the long-term cycling stability of the material.^27–29 In addition, the narrow interlayer spacing, inherent poor conductivity, and limited active sites of NH₄V₄O₁₀ lead to poor electrochemical kinetic performance.^21,30,31 As a result, the specific capacity (380.3 mA h g⁻¹ at 0.1 A g⁻¹) and cycling stability (only 33% capacity retention after 100 cycles) of the pure NH₄V₄O₁₀ cathodes are generally poor.²⁵ Several effective strategies have been proposed to enhance the structural stability and improve the kinetic behavior of NH₄V₄O₁₀, including morphological modulation,²⁴ defect engineering,²⁴ and pre-intercalation.³² As an illustration, Huang et al. synthesized 3D nanoflower NH₄V₄O₁₀ by a microwave-assisted method, successfully increasing the reversible capacity and improving the electrochemical kinetics.²⁴ He et al. synthesized oxygen-deficient NH₄V₄O_10−x·nH₂O microspheres, in which the oxygen vacancies induced in the crystal lattice lower the Zn²⁺ diffusion energy barriers and enable rapid diffusion.²⁶ Wang et al. introduced Mg²⁺ into the interlayer of NH₄V₄O₁₀ to stabilize the layered structure by enhancing the ionic bonding.³² In addition, organic molecules can be employed as interlayer pillars to expand the interlayer spacing and enhance the zinc storage performance of layered vanadium-based materials. For example, Xu et al. designed C₃N₄ column-supported NH₄V₄O₁₀, successfully extending the interlayer spacing, inducing abundant oxygen vacancies, and thus promoting the Zn²⁺ ion (de)intercalation kinetics and ionic conductivity.³³ Surprisingly, metal cations and organic molecules can be simultaneously introduced into the interlayers to synergistically extend the layer spacing of the layered material, increase the strength of ionic bonding, and modulate the crystal structure. For example, Zhao et al. co-inserted Na⁺ ions and polyaniline into the interlayers of NH₄V₄O₁₀ (NaNVO-PANI) nano-arrays to expand the interlayer spacing, improving the electronic conductivity of the nanoparticles and lowering the diffusion barrier of Zn²⁺.³⁴ On the other hand, we note that the electronegativity of K is smaller than that of V, so it can form stronger chemical bonds with oxygen, thus alleviating the structural deformation during cycling and effectively improving the stability of the lamellar structure.^32,35 Furthermore, the low-valent doping of K⁺ may induce oxygen vacancies.³⁶ In addition, the larger molecular volume of ethylene glycol can significantly expand the interlayer spacing of NH₄V₄O₁₀ to obtain larger ion diffusion channels.³⁷ Simultaneously, the strong electronegative O atoms with lone pair of electrons in ethylene glycol can establish O–H⋯O hydrogen bonding interactions with the V–O layer, essentially improving the structural stability of NH₄V₄O₁₀ through reconstruction of the internal structure,²⁵ and ethylene glycol has the potential to decrease the solvent's polarity in a reaction and influence the nucleation and growth of the material through surface-energy-driven recrystallization, thereby optimizing the microscopic morphology of NH₄V₄O₁₀.^12,38 Consequently, it is reasonable to anticipate that co-inserting K⁺ ions and ethylene glycol into the interlayer of NH₄V₄O₁₀ may significantly enhance the ionic bonding, improve the crystal structure optimization, control the microscopic morphology, and thus unlock the zinc storage performance of the NH₄V₄O₁₀ cathode.

Herein, we constructed NH₄V₄O₁₀ nanoflowers co-intercalated with K⁺ ions and ethylene glycol (K, EG-NVO) as an ultrahigh-capacity cathode for AZIBs. It is found that the co-intercalation of K⁺ and ethylene glycol synergistically regulates the interlayer spacing, oxygen defects and microscopic morphology: (1) the embedded K⁺ ions and EG replace part of the NH₄⁺, induce a higher content of oxygen defects, and enhance the ionic bond strength in NH₄V₄O₁₀; (2) the co-intercalation of K⁺ ions and ethylene glycol successfully enlarges the layer spacing of NH₄V₄O₁₀ to 11.5 Å, ensuring large Zn²⁺ diffusion channels and accelerating the reaction kinetics; (3) the introduction of K⁺ ions and ethylene glycol weakens the crystallinity of NH₄V₄O₁₀ and leads to the transformation of ribbons to nanoflowers self-assembled from ultrathin nanosheets, facilitating efficient redox reactions; and (4) the band gap is decreased by 0.2 eV after the co-intercalation of K⁺ and EG, enhancing the electronic conductivity and favoring the diffusion of Zn²⁺. Accordingly, K, EG-NVO delivers the ultra-high capacity of 614.1 mA h g⁻¹ at 0.5 A g⁻¹, outstanding rate performance of 472.9 mA h g⁻¹ at 10 A g⁻¹ and high-capacity retention of up to 89% after 2000 cycles, demonstrating successful unlocking of Zn storage properties of NH₄V₄O₁₀ after the co-intercalation of K⁺ ions and ethylene glycol.

2. Results and discussion

In this work, all the samples are synthesized by a one-step hydrothermal method. Potassium sulfate, ethylene glycol, and potassium sulfate/ethylene glycol are added to the solution to obtain K⁺-intercalated NH₄V₄O₁₀ (K-NVO), ethylene glycol-intercalated NH₄V₄O₁₀ (EG-NVO) and K, EG-NVO materials, respectively. Detailed experimental procedures for each material are presented in the ESI.† XRD spectra of NVO, K-NVO, EG-NVO, and K, EG-NVO materials match well with the monoclinic structure of NH₄V₄O₁₀ (JCPDS: 31-0075), and no impurity peaks are detected for all the materials (Fig. 1a). It is evident from Fig. 1a that the crystallinities of K-NVO, EG-NVO and K, EG-NVO are significantly weaker compared to that of NVO, providing more Zn²⁺ migration pathways. The strong signal at 2θ = 9.20° in Fig. 1b corresponds to the (001) crystal plane with a layer spacing of 9.6 Å in NVO. When K⁺ ions are introduced into the interlayer of NVO, the diffraction peak of the (001) crystal plane is shifted to 9.26°, and the corresponding layer spacing is reduced to 9.5 Å. This is due to the lower electronegativity of K⁺ (0.82) than V and high binding energy with the oxygen atoms in the V–O interlayer, leading to a decrease in lattice spacing. After the introduction of ethylene glycol, the peak of the (001) crystal plane is noticeably shifted to 7.4°, indicating that the lattice spacing of the material is increased to 11.9 Å. However, when K⁺ ions and ethylene glycol are co-inserted between the NVO layers, the diffraction peak of the (001) crystal plane is slightly increased to 7.7°, corresponding to an interlayer spacing of 11.5 Å. When Zn²⁺ ions diffuse between the layers, a larger layer spacing is favorable to reducing its coulombic interaction with the layers. Meanwhile, the strong ionic bonds formed between the K⁺ ions and the V–O layer ensure the stability of the lamellar structure. Fourier transform infrared (FT-IR) spectroscopy analysis confirms the similarity of the crystal structures of NVO, K-NVO, EG-NVO and K, EG-NVO (Fig. 1c). The peaks observed at around 1623 cm⁻¹ indicate the bending vibrations of water, providing evidence for the presence of structural water molecules in the materials. The peaks near 1410 cm⁻¹ correspond to the bending vibrations of the N–H bond, indicating the presence of NH₄⁺. Moreover, the peaks located near 1003, 766, and 520 cm⁻¹ are attributed to different vibrational modes of V=O, V–O–V, and V–O in NVO, K-NVO, EG-NVO, and K, EG-NVO, respectively, where the peak at 520 cm⁻¹ is associated with the bending vibrational mode of V–O. The blue shift of the V–O peak is observed in K-NVO, EG-NVO and K, EG-NVO, which is attributed to the oxygen vacancies induced by the introduction of K⁺ ions and EG. The peak in NVO located at 152 cm⁻¹ in the Raman spectrum of Fig. 1d is affected by the bending vibration of the V–O chain. Partial hydrogen bonding between the N–H and V–O layers is broken due to the partial substitution of the K⁺ ions and EG molecules for some of the NH₄⁺ ions, resulting in a shift of the peaks associated with the V–O bending vibrations to lower wave numbers. From the XPS spectra of C 1s in Fig. 1e, K, EG-NVO shows a distinct new peak at 292.5 eV compared to the blank NVO, which is related to the intercalated ethylene glycol.³⁹ The successful insertion of K⁺ ions is verified by the XPS spectrum of K 2p in the K, EG-NVO sample (Fig. 1f). In addition, the contents of K⁺ ions, ethylene glycol molecules and structured water molecules in the NVO-based samples are further characterized by ICP tests, thermogravimetric analysis and elemental analysis, respectively (Tables S1–S3†). Electron paramagnetic resonance (EPR) spectroscopy studies further verify the presence of oxygen vacancies in the materials. The symmetric peaks with g = 1.995 are present for both NVO and K, EG-NVO materials (Fig. 1g). Apparently, the signal peak is much stronger in K, EG-NVO than in NVO, indicating that more oxygen vacancies are formed in K, EG-NVO. In addition, the O 1s XPS spectra of NVO and K, EG-NVO can be fitted to three peaks, where the characteristic peaks at 530 eV, 530.9 eV, and 532.2 eV represent the V–O bonds, oxygen defects, and H₂O molecules, respectively (Fig. 1h). The proportion of the peak with oxygen vacancies in K, EG-NVO is significantly higher than that in NVO, which further confirms that more oxygen vacancies are formed in K, EG-NVO. In the XPS profile of V 2p (Fig. 1i), the peaks located at 516.1/523.9 eV and 517.5/525.1 eV correspond to V⁴⁺ and V⁵⁺, respectively. Based on the fitted peak areas of V 2p, it is clear that the ratio of V⁴⁺/V⁵⁺ in K, EG-NVO is higher than that in NVO, which is due to the fact that the inserted K⁺ ions and EG molecules lead to an increase in the content of oxygen vacancy, and thus some of the V⁵⁺ is reduced to V⁴⁺.

Fig. 1 Phase and spectral characterization profiles of various cathode materials. (a and b) XRD patterns. (c) FTIR spectra. (d) Raman spectra. XPS spectra of (e) C 1s and (f) K 2p. (g) EPR spectra. XPS spectra of (h) O 1s and (i) V 2p.

Fig. 2 and S1–S3† show the morphology and microstructure of NVO, K-NVO, EG-NVO, and K, EG-NVO. According to Fig. S1,† NVO is composed of strips of irregular sizes. Interestingly, K-NVO, EG-NVO and K, EG-NVO possess nanoflower morphologies self-assembled from ultrathin nanosheets after the introduction of K⁺ and ethylene glycol (Fig. 2a, b, S2 and S3†). TEM images of K, EG-NVO further confirm its ultrathin nanosheet morphology (Fig. 2c and d). This may be due to the fact that the addition of K⁺ and EG during the hydrothermal reaction modulates the nucleation and growth of the material through surface energy-driven recrystallization, leading to the change in micromorphology. Obviously, compared with NVO, K, EG-NVO possesses a larger specific surface area and richer active sites, which are favorable for the complete penetration of the electrolyte and the transfer of electrons and ions. From Fig. S4,† the contact angles of the electrolyte with NVO, K-NVO, EG-NVO and K, EG-NVO electrodes are 114.40°, 110.43°, 110.16° and 94.57°, respectively, which further confirm that the introduction of K⁺ and EG is more favorable for electrolyte infiltration. The larger specific surface area of K, EG-NVO (15.911 m² g⁻¹) than that of blank NVO (14.403 m² g⁻¹) is further confirmed using the nitrogen adsorption–desorption isotherms of the materials, and both materials are mesoporous (Fig. S5 and S6.†). As shown in Fig. 2e, K, EG-NVO exhibits various lattice defects caused by oxygen vacancies and shows single-crystal properties. The lattice fringes in Fig. 2f exhibit clear patterns, with lattice spacings of 0.196 nm and 0.191 nm, representing the (−205) and (−314) crystal planes of K, EG-NVO, respectively. Its elemental mapping confirms that C, K, N, O and V elements are uniformly distributed in the nanosheets, suggesting that K⁺ ions and EG molecules are successfully inserted into the interlayer of NVO (Fig. 2g).

Fig. 2 (a and b) SEM and (c and d) TEM images of K, EG-NVO. (e and f) HR-TEM and SAED images and (g) TEM-EDS elemental mapping patterns of K, EG-NVO.

Electrochemical tests are performed on the NVO, K-NVO, EG-NVO, and K, EG-NVO cathodes, and the results are shown in Fig. 3 and S7–S9.† The CV curves of NVO and K, EG-NVO for the first three cycles at 0.3–1.6 V and 0.1 mV s⁻¹ are shown in Fig. 3a. The CV curves all show four similar redox peaks at 1.09/0.93 V, 1.02/0.80 V, 0.52/0.58 V, and 0.39/0.38 V, which correspond to the multistep intercalation/deintercalation behavior of Zn²⁺ during charging and discharging. In this case, the redox reaction of V⁴⁺/V³⁺ occurs around ∼0.5 V, while the peak at the higher voltage of ∼1 V corresponds to the redox reaction of V⁵⁺/V⁴⁺. The small redox peak located at 1.39/1.31 V may be related to the intercalation/deintercalation of Zn²⁺ in the Zn₃(OH)₂V₂O₇·2H₂O phase formed during discharging. Apparently, the peaks of K, EG-NVO corresponding to the Zn₃(OH)₂V₂O₇·2H₂O phase are much smaller, and the CV curves are highly overlapping compared with those of NVO, suggesting more excellent cycling stability. As shown in Fig. 3b and c and S7,† the reactivity of the K, EG-NVO cathode is much higher than that of NVO. The first discharge capacity of K, EG-NVO is 376.1 mA h g⁻¹ at 0.5 A g⁻¹, which is dramatically increased to 614.1 mA h g⁻¹ after cycling, which is much higher than that of NVO (409.6 mA h g⁻¹), K-NVO (525.9 mA h g⁻¹) and EG-NVO (513.5 mA h g⁻¹). In addition, the K, EG-NVO electrode has a more remarkable capacity retention of 97% after 50 cycles than the others. The discharge capacities of the K, EG-NVO cathode are 604, 594.5, 552.1, 466.5, 433.1, and 315.6 mA h g⁻¹ at 0.5, 1, 2, 4, 5, and 10 A g⁻¹, respectively, and they return to 424.8, 449.8, 521.6, 559.7 and 565.9 mA h g⁻¹, respectively, when current densities are restored to 5, 4, 2, 1, and 0.5 A g⁻¹ (Fig. 3d). Apparently, compared to the NVO cathode (444.9, 408.2, 322.2, 230, 192.2, and 81.5 mA h g⁻¹ at 0.5, 1, 2, 4, 5, and 10 A g⁻¹, respectively) and other samples (K-NVO: 549, 517.7, 467.8, 400.8, 373.4, and 249.9 mA h g⁻¹ at 0.5, 1, 2, 4, 5, and 10 A g⁻¹, respectively; EG-NVO: 546, 512.2, 380.4, 221.1, 175.9, and 69.6 mA h g⁻¹ at 0.5, 1, 2, 4, 5, and 10 A g⁻¹, respectively), the rate performance of the K, EG-NVO electrode is much better at different current densities (Fig. 3d and S8†). In addition, the highest discharge-specific capacity of the K, EG-NVO cathode is as high as 472.9 mA h g⁻¹ at a high current density of 10 A g⁻¹, and even after 2000 cycles K, EG-NVO exhibits a high capacity of 419.7 mA h g⁻¹ with an excellent capacity retention rate of 89%. In contrast, the first discharge capacity of NVO is 90.2 mA h g⁻¹ at 10 A g⁻¹, which is increased to 122.4 mA h g⁻¹ after 2000 cycles, indicating that the activation process of NVO is much longer, which is not conducive to practical applications (Fig. 3e). Meanwhile, K-NVO and EG-NVO exhibit low capacities of 281.4 and 294.7 mA h g⁻¹ with capacity retentions of 90% and 77% after 2000 cycles, respectively (Fig. S9†). Fig. 3f and Table S4† summarize the electrochemical performance of recently reported various advanced V-based cathode materials for AZIBs, reflecting that the K, EG-NVO cathode is comparable or superior to the state-of-the-art cathodes for AZIBs.^{32,34,40–47}

Fig. 3 Electrochemical testing of different cathode electrodes. (a) CV curves at 0.1 mV s⁻¹. (b) GCD curves at 0.5 A g⁻¹. (c) Electrochemical properties at 0.5 A g⁻¹. (d) Rate performances. (e) Cycling performance at 10 A g⁻¹. (f) Comparison of rate performances.

The electrochemical kinetics of the K, EG-NVO and NVO cathodes are investigated through CV curves at different scan rates (Fig. 4a–f). The shapes of the CV curves remain essentially the same at different scanning speeds. The equation i = aV^b can be used to represent the relationship between the maximum current (i) and the scanning speed (V), where both a and b values are variation parameters. When the value of b is close to 0.5, the charge storage process is dominated by diffusion-controlled processes, while when b is close to 1, the charge storage process is dominated by pseudocapacitance-controlled processes. The b-values of the redox peaks 1–5 of K, EG-NVO and NVO are 0.81, 0.99, 0.73, 0.66, and 0.88 and 0.67, 0.77, 0.71, 0.51, and 0.56, respectively, proving that their charge storage processes are controlled by a combination of diffusive and pseudocapacitive behaviors (Fig. 4b and e). In addition, the equation i = k₁v + k₂v^1/2 can be used to approximate the diffusion (k₂v^1/2) and pseudocapacitance (k₁v) control contributions. Fig. 4c and f show that when the scan rate is increased from 0.1 mV s⁻¹ to 0.25 mV s⁻¹, the pseudocapacitance contribution of the K, EG-NVO anode is increased from 81% to 95%, whereas the pseudocapacitance contribution of the NVO anode is only increased from 54% to 65%, showing that the kinetics of K, EG-NVO is better than that of NVO.⁴⁸Fig. 4g–h show the GITT curves and diffusion coefficients for K, EG-NVO and NVO. During charging and discharging, the D_Zn²⁺ values of the K, EG-NVO electrodes are in the range of 10⁻¹⁰ to 10⁻⁸ cm² s⁻¹, which are significantly higher than those of NVO (10⁻¹² to 10⁻⁹ cm² s⁻¹), providing more evidence for faster charge transfer kinetics of K, EG-NVO than NVO.

Fig. 4 (a) CV curves, (b) b-values and (c) pseudocapacitance contributions of K, EG-NVO at different scanning speeds. (d) CV curves, (e) b-values and (f) pseudocapacitance contributions of NVO at different scanning speeds. GITT curves at 0.1 A g⁻¹ and Zn²⁺ diffusion coefficient of (g) K, EG-NVO and (h) NVO.

The reasons for excellent Zn²⁺ storage on the K, EG-NVO electrode with inserted K⁺ ions and ethylene glycol are further analyzed by density functional theory (DFT) calculations.⁴⁹ The crystal structures of the optimized NVO and K, EG-NVO are shown in Fig. S10.† The projected density of states (PDOS) calculations shown in Fig. 5a and b indicate that the K, EG-NVO cathode shows new electronic states near the Fermi energy level, and the band gap is significantly reduced from 1.02 to 0.82 eV compared to that of NVO, which greatly improves the electron transport behavior. Fig. 5c, d, f and g and S11† model the structure of each sample by simulating the insertion of Zn²⁺ ions into the structural framework. From the binding energy calculations of Zn²⁺ ions for each material, it can be seen that K, EG-NVO exhibits the lowest binding energy (−3.621 eV) compared to other materials (NVO: −3.841 eV, K-NVO: −3.832 eV, and EG-NVO: −3.624 eV), suggesting that the K, EG-NVO electrode is more favorable for the insertion and extraction of Zn²⁺, resulting in significant improvement in the ion transport behavior (Fig. 5e). In addition, the possible Zn²⁺ migration paths are selected (Fig. 5h–k) for the calculation of diffusion energy barriers. From the calculation results, the co-intercalation of K⁺ ions and ethylene glycol significantly reduces the diffusion barrier of NVO from 0.29 eV to 0.13 eV, indicating that co-intercalation is more favorable for the diffusion of Zn²⁺ ions. The UV-vis spectra shown in Fig. S12a and b† reveal that the optical band gaps of NVO and K, EG-NVO are 1.94 and 1.82 eV, respectively, which further verify that the co-intercalation of K⁺ ions and EG leads to the decrease of the band gap of NVO.⁵⁰ Ultraviolet photoelectron spectroscopy (UPS) is used to study the Fermi level spectra of the samples and the variation of the surface electronic structure. The calculated valence band maximum energy levels (EVBM) of NVO and K, EG-NVO are 16.62 and 16.32 eV, respectively (Fig. S12c and d†). The calculated work functions (FoM) of 4.6 and 4.9 eV for NVO and K, EG-NVO, respectively, suggest that the electronic structure of K, EG-NVO is slightly changed due to the co-intercalation of K⁺ ions and EG.⁵¹

Fig. 5 Partial density of states (PDOS) of (a) NVO and (b) K, EG-NVO. Geometric structures of (c) NVO, (d) K-NVO, (f) EG-NVO, and (g) K, EG-NVO with inserted zinc ions. (e) Binding energies for Zn²⁺ ions in NVO, K-NVO, EG-NVO, and K, EG-NVO. Simulated migration paths of Zn²⁺ in (i) NVO and (j) K, EG-NVO. Calculated Zn²⁺ diffusion barriers in (h) NVO and (k) K, EG-NVO cathodes.

To investigate the charge storage mechanism, the phase changes of the K, EG-NVO cathode during the initial two charging and discharging cycles in 2 M Zn(OTF)₂ electrolyte are examined by in situ XRD (Fig. 6a–d). To explore more clearly, the peaks near 50° are enlarged in Fig. 6c. It can be observed that the peaks located at 48.7° and 59.9° are obviously shifted towards lower angles during discharge, indicating that the layer spacing increases as Zn²⁺ is embedded in the interlayer of K, EG-NVO, whereas both peaks are shifted towards a higher angle during charging, implying that Zn²⁺ ions are detached from the K, EG-NVO cathode, leading to a decrease in layer spacing. The diffraction peak around 33° is magnified in Fig. 6d. The voltage is gradually reduced to 0.3 V during discharge, and a new diffraction peak appears around 33.6°, which is related to the formation of Zn₃(OH)₂V₂O₇·2H₂O (JCPDS: 50-0570) by-products, whereas upon reaching a charge of 1.6 V, the peak associated with Zn₃(OH)₂V₂O₇·2H₂O disappears. Similar appearance and disappearance of the peaks are observed in the subsequent cycles, suggesting good reversibility of the K, EG-NVO cathode during charging and discharging. To further explore the appearance/disappearance of byproducts, the SEM tests of the cathode after charge/discharge cycling are performed (Fig. S13–S17†). When discharged to 0.3 V, noticeable larger-sized flakes appear compared to the initial state (Fig. S13†), covering the surface of the cathode material and corresponding to the generated Zn₃(OH)₂V₂O₇·2H₂O (Fig. S14 and S16†), whereas when charged to 1.6 V, the flakes disappear, corresponding to the disappearance of by-products (Fig. S15 and S17†).

Fig. 6 (a–d) In situ XRD patterns of K, EG-NVO. Ex situ XPS spectra for (e) Zn 2p, (f) V 2p, and (g) O 1s of K, EG-NVO. (h and i) SEM and (j and k) HR-TEM images of the K, EG-NVO electrode after cycling.

As shown in Fig. 6e–g, ex situ XPS is used to analyze the changes in V valence state, Zn and H₂O content in the K, EG-NVO cathode during charging and discharging. As shown in Fig. 6e, no obvious Zn 2p signal peaks are detected in the pristine state. Two strong characteristic peaks located at 1046.3 eV and 1023.2 eV are detected in the fully discharged state, indicating the successful insertion of Zn²⁺ ions into the K, EG-NVO cathode, whereas the intensity of the Zn peaks decreases when charging up to 1.6 V, corresponding to the extraction of Zn²⁺ ions from the host material. However, the signal peaks of Zn are still present in the fully charged state, which is due to the presence of residual electrolyte and Zn²⁺ ions that are not fully extracted. The XPS spectrum of V 2p is shown in Fig. 6f. Compared with the initial state, the area of the peak corresponding to V⁵⁺ decreases significantly, the area of the peak of V⁴⁺ increases when discharged to 0.3 V, and a new peak of V³⁺ appears, suggesting that the oxidation state of V decreases significantly with the insertion of Zn ions. When charged to 1.6 V, the peak of V³⁺ disappears and the peak area of V⁵⁺ increases, indicating the reversible redox reaction of V. As shown in the XPS spectrum of O 1s (Fig. 6g), the proportion of the peaks associated with water molecules increases in the discharged state, indicating that water molecules and Zn²⁺ ions are co-inserted, and then the area of the H₂O peaks decreases with the detachment of Zn²⁺ ions. SEM and TEM tests are performed after charge/discharge cycles to better verify the cycling stability of the cathode material. As shown in Fig. 6h–k, the K, EG-NVO material after charge/discharge cycling still exhibits an ultrathin lamellar morphology and clear lattice stripes, indicating its excellent structural stability.

Based on the above in situ XRD, ex situ XPS, SEM and TEM characterizations studies, the storage mechanism of Zn²⁺ ions during charging and discharging is determined, and the schematic diagram of the phase transformation is shown in Fig. 7. The electrode reactions involved are as follows:

Fig. 7 Schematic diagram of the Zn storage mechanism in K, EG-NVO.

First discharge on the cathode:

K, EG-NH₄V₄O₁₀ + nZn²⁺ + 2ne⁻ → K, EG -Zn_xNH₄V₄O₁₀ + Zn₃V₂O₇(OH)₂·2H₂O

The following cycle on the cathode:

K, EG-Zn_xNH₄V₄O₁₀ + Zn₃V₂O₇(OH)₂·2H₂O – yZn²⁺ – 2ye⁻ ↔ K, EG-Zn_x−zNH₄V₄O₁₀

3. Conclusion

In summary, we have synthesized NH₄V₄O₁₀ nanoflowers co-intercalated with K⁺ ions and ethylene glycol as a high-performance cathode for AZIBs using a simple one-step solvothermal method. Compared to NVO, the K, EG-NVO electrode exhibits an ultrahigh capacity of 614.1 mA h g⁻¹ at 0.5 A g⁻¹, a superior rate performance of 472.9 mA h g⁻¹ at 10 A g⁻¹, and an outstanding cycling stability with capacity retention up to 89% after 2000 cycles at 10 A g⁻¹. The greatly enhanced electrochemical properties of the K, EG-NVO electrode can be attributed to the synergistic modulation of interlayer spacing, oxygen defects, and microscopic morphology resulting from the co-intercalation of K⁺ ions and ethylene glycol. It is found that the co-intercalation of K⁺ ions and ethylene glycol into NVO enlarges the interlayer distance to 11.5 Å, providing larger interlayer channels for (de)intercalation of Zn²⁺. Meanwhile, the co-intercalation of K⁺ ions and ethylene glycol weakens the crystallinity of the material, causes the transformation of the microscopic morphology of the material into nanoflowers self-assembled from ultrathin nanosheets and increases the content of oxygen defects, synergistically resulting in more active sites and promoting the transport of ions and electrons. In addition, DFT calculations further confirm that the introduction of K⁺ ions and ethylene glycol contributes to the enhancement of the electronic conductivity of the cathode material as well as the reduction of the diffusion energy barrier of Zn²⁺ ions. In this study, a feasible method to synergistically modulate the layer spacing, crystal structure and microscopic morphology of layered V-based materials by pre-embedding metal cations and organic molecules is proposed to unlock the Zn storage behavior in layer-structured V-based cathodes for high-performance AZIBs.

Data availability

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

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Sichuan Science and Technology Program (2023NSFSC0439). We gratefully acknowledge HZWTECH for providing computation facilities. The authors also appreciate the Shiyanjia Lab (http://www.shiyanjia.com) for the SEM tests, eceshi (http://www.eceshi.cn) for the TEM tests and Beijing Nordson Rongke Technology Co., Ltd (http://www.kexingtest.com) for the BET tests.

References

1. Q. Pang, W. He, X. Yu, S. Yang, H. Zhao, Y. Fu, M. Xing, Y. Tian, X. Luo and Y. Wei, Appl. Surf. Sci., 2021, 538, 148043.
2. S. S. Zhang, Energy Environ. Mater., 2022, 5, 1005–1007.
3. Y. J. Kim, S. Y. Ko, S. Kim, K. M. Choi and W. H. Ryu, Small, 2023, 19, 2206561.
4. M. Kim, E. Seok, J. Park, S. Lee, H. Kang, M. Ku, K. Y. Chung, H. Jung and W. Choi, Chem. Eng. J., 2022, 450, 137939.
5. Z. Liu, Z. Guo, L. Fan, C. Zhao, A. Chen, M. Wang, M. Li, X. Lu, J. Zhang, Y. Zhang and N. Zhang, Adv. Mater., 2023, 36, 2305988.
6. Y. Li, J. Chen, L. Su, X. Zhang, Q. Zheng, Y. Huo and D. Lin, J. Colloid Interface Sci., 2023, 652, 440–448.
7. Y. Li, Y. Liu, J. Chen, Q. Zheng, Y. Huo, F. Xie and D. Lin, Chem. Eng. J., 2022, 448, 137681.
8. Q. Hu, J. Hou, Y. Liu, L. Li, Q. Ran, J. Mao, X. Liu, J. Zhao and H. Pang, Adv. Mater., 2023, 35, 2303336.
9. H. Cao, X. Huang, Y. Li, Y. Liu, Q. Zheng, Y. Huo, R. Zhao, J. Zhao and D. Lin, Chem. Eng. J., 2023, 455, 140538.
10. X. Zhang, J. Chen, H. Cao, X. Huang, Y. Liu, Y. Chen, Y. Huo, D. Lin, Q. Zheng and K. H. Lam, Small, 2023, 19, 2303906.
11. Y. Zhang, Z. Liu, X. Li, L. Fan, Y. Shuai and N. Zhang, Adv. Energy Mater., 2023, 13, 2302126.
12. J. Chen, L. Su, X. Zhang, Y. Chen, P. Wang, Q. Zheng and D. Lin, ACS Sustainable Chem. Eng., 2023, 11, 12467–12476.
13. J. Chen, Y. Zhai, Y. Li, X. Zhang, X. Zhang, Y. Chen, Y. Zeng, X. Wu, Q. Zheng, K. H. Lam, X. Tan and D. Lin, Small, 2024, 20, 2309412.
14. L. Hu, Z. Wu, C. Lu, F. Ye, Q. Liu and Z. Sun, Energy Environ. Sci., 2021, 14, 4095–4106.
15. Y. Bai, H. Zhang, W. Liang, C. Zhu, L. Yan and C. Li, Small, 2024, 20, 2306111.
16. Y. Bai, L. Luo, W. Song, S. Man, H. Zhang and C. M. Li, Adv. Sci., 2024, 11, 2308668.
17. X. Ma, X. Cao, M. Yao, L. Shan, X. Shi, G. Fang, A. Pan, B. Lu, J. Zhou and S. Liang, Adv. Mater., 2022, 34, 2105452.
18. K. Wang, H. Li, G. Guo, L. Zheng, S. Passerini and H. Zhang, ACS Energy Lett., 2023, 8, 1671–1679.
19. Y. Bai, Y. Qin, J. Hao, H. Zhang and C. M. Li, Adv. Funct. Mater., 2023, 34, 2310393.
20. Y. Bai, H. Zhang, H. Song, C. Zhu, L. Yan, Q. Hu and C. M. Li, Nano Energy, 2024, 120, 109090.
21. D. He, Y. Peng, Y. Ding, X. Xu, Y. Huang, Z. Li, X. Zhang and L. Hu, J. Power Sources, 2021, 484, 229284.
22. R. Li, F. Xing, T. Li, H. Zhang, J. Yan, Q. Zheng and X. Li, Energy Storage Mater., 2021, 38, 590–598.
23. T.-F. Yi, L. Qiu, J.-P. Qu, H. Liu, J.-H. Zhang and Y.-R. Zhu, Coord. Chem. Rev., 2021, 446, 214124.
24. Q. Li, X. Rui, D. Chen, Y. Feng, N. Xiao, L. Gan, Q. Zhang, Y. Yu and S. Huang, Nano-Micro Lett., 2020, 12, 1–12.
25. B. Tang, J. Zhou, G. Fang, F. Liu, C. Zhu, C. Wang, A. Pan and S. Liang, J. Mater. Chem. A, 2019, 7, 940–945.
26. T. He, Y. Ye, H. Li, S. Weng, Q. Zhang, M. Li, T. Liu, J. Cheng, X. Wang, J. Lu and B. Wang, Mater. Today, 2021, 43, 53–61.
27. X. Wang, Z. Zhang, M. Huang, J. Feng, S. Xiong and B. Xi, Nano Lett., 2021, 22, 119–127.
28. J. Wang, F. Zheng, M. Li, J. Wang, D. Jia, X. Mao, P. Hu, Q. Zhen and Y. Yu, Chem. Eng. J., 2022, 446, 136922.
29. G. Li, L. Sun, S. Zhang, C. Zhang, H. Jin, K. Davey, G. Liang, S. Liu, J. Mao and Z. Guo, Adv. Funct. Mater., 2023, 19, 2301291.
30. H. Jiang, Y. Zhang, Y. Liu, J. Yang, L. Xu, P. Wang, Z. Gao, J. Zheng, C. Meng and Z. Pan, J. Mater. Chem. A, 2020, 8, 15130–15139.
31. X. Wang, Y. Wang, Y. Jiang, X. Li, Y. Liu, H. Xiao, Y. Ma, Y. Y. Huang and G. Yuan, Adv. Funct. Mater., 2021, 31, 2103210.
32. X. Wang, Y. Wang, A. Naveed, G. Li, H. Zhang, Y. Zhou, A. Dou, M. Su, Y. Liu, R. Guo and C. C. Li, Adv. Funct. Mater., 2023, 33, 2306205.
33. Y. Xu, G. Fan, P. X. Sun, Y. Guo, W. Yangyang, X. Gu, L. Wu and L. Yu, Angew. Chem., Int. Ed., 2023, 62, e202303529.
34. S. Zhao, S. Wang, J. Guo, L. Li, C. Li, Y. Sun, P. Xue, D. Wu, L. Wei, Y. Wang and Q. Zhang, Adv. Funct. Mater., 2023, 33, 2305700.
35. Z. Feng, Y. Zhang, J. Sun, Y. Liu, H. Jiang, M. Cui, T. Hu and C. Meng, Chem. Eng. J., 2022, 433, 133795.
36. J. J. Carey and M. Nolan, J. Mater. Chem. A, 2017, 5(30), 15613–15630.
37. Z. Lv, Y. Tan, Y. Kang, J. Yang, X. Cheng, W. Meng, Y. Zhang, C. C. Li, J. Zhao and Y. Yang, Sci. China: Chem., 2023, 66, 1537–1548.
38. S. Zhong, Z. Zou, S. Lv, S. Zhang, J. Geng, J. Meng, X. Liu, F. Liang and J. Rao, J. Alloys Compd., 2022, 903, 163845.
39. W. He, Z. Lin, K. Zhao, Y. Li, C. Meng, J. Li, S. Lee, Y. Wu and X. Hao, Adv. Mater., 2020, 34, 2203920.
40. W. He, C. Meng, Z. Ai, D. Xu, S. Liu, Y. Shao, Y. Wu and X. Hao, Chem. Eng. J., 2023, 454, 140260.
41. Q. Zong, Y. Zhuang, C. Liu, Q. Kang, Y. Wu, J. Zhang, J. Wang, D. Tao, Q. Zhang and G. Cao, Adv. Energy Mater., 2023, 13, 2301480.
42. J. Guo, J. Liu, W. Ma, Z. Sang, L. Yin, X. Zhang, H. Chen, J. Liang and D. A. Yang, Adv. Funct. Mater., 2023, 33, 2302659.
43. J. Yang, J. Li, Y. Li, Z. Wang, L. Ma, W. Mai, M. Xu and L. Pan, Chem. Eng. J., 2023, 468, 143600.
44. W. Wang, R. Hu, C. Zhang, Y. Tao, L. Ran, Y. Li, Y. Ouyang and J. Yan, J. Mater. Chem. A, 2023, 11, 8224–8234.
45. W. Deng, Z. Xu, G. Li and X. Wang, Small, 2023, 19, 2207754.
46. M. Wu, C. Shi, J. Yang, Y. Zong, Y. Chen, Z. Ren, Y. Zhao, Z. Li, W. Zhang, L. Wang, X. Huang, W. Wen, X. Li, X. Ning, X. Ren and D. Zhu, Adv. Mater., 2024, 36, 2310434.
47. L. Wang, K.-W. Huang, J. Chen and J. Zheng, Sci. Adv., 2019, 5, eaax4279.
48. N. Liu, X. Wu, L. Fan, S. Gong, Z. Guo, A. Chen, C. Zhao, Y. Mao, N. Zhang and K. Sun, Adv. Mater., 2020, 32, 1908420.
49. P. E. Blöchl, Phys. Rev. B, 1994, 50, 17953–17979.
50. R. R. Katzbaer, F. M. dos Santos Vieira, I. Dabo, Z. Mao and R. E. Schaak, J. Am. Chem. Soc., 2023, 145, 6753–6761.
51. Y. Xiao, C. Feng, J. Fu, F. Wang, C. Li, V. F. Kunzelmann, C.-M. Jiang, M. Nakabayashi, N. Shibata, I. D. Sharp, K. Domen and Y. Li, Nat. Catal., 2020, 3, 932–940.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta04112g

‡ These authors contributed equally.

---