Layered hydrated vanadium oxide as highly reversible intercalation cathode for aqueous Zn‐ion batteries

Abstract

Aqueous Zn‐ion batteries (ZIBs) hold great potential in large‐scale energy storage systems due to the merits of low‐cost and high safety. However, the unstable structure of cathode materials and sluggish (de)intercalation kinetics of Zn²⁺ pose challenges for further development. Herein, highly reversible aqueous ZIBs are constructed with layered hydrated vanadium oxide as a cathode material. The electrochemical performances are further tested with the optimized electrolyte of 3M Zn(CF₃SO₃)₂ and a cut‐off voltage of 0.4 to 1.3 V, exhibiting a remarkable capacity of 290 mAh g⁻¹ at 0.5 A g⁻¹, and long‐term cycling stability at high current density. Furthermore, the Zn²⁺ storage mechanism of V₃O₇⋅H₂O is recognized as a highly reversible (de)intercalation process with good structural stability, implying the potential application in the field of large‐scale energy storage.

INTRODUCTION

With the continuous consumption of fossil resources and booming development of new energy system, exploring potential secondary battery for large‐scale energy storage is an urgent desire.^1‐3 Among the existing battery systems, lithium‐ion batteries (LIBs) have predominated the global market for decades. However, their drawbacks such as high‐cost, poor safety, and scarce lithium resources, severely confine the further development of LIBs.^4‐6 Nowadays, owing to the advantages of good specific capacity, low‐cost, high‐safety, and similar working mechanism to LIBs, aqueous Zn‐ion batteries (ZIBs) have attracted tremendous interests in the field of large‐scale energy storage.^7,8 As a result of its relatively low redox potential (−0.76 V vs standard hydrogen electrode) and high overpotential of hydrogen evolution, Zn metal presents a highly reversible stripping and depositing behavior in aqueous electrolyte.^9‐11

Cathode materials with the tunneled or layered structure are the hot research points in aqueous ZIBs,^12‐15 mainly including vanadium‐based oxides,^16‐20 and manganese‐based oxides.^21‐24 Notably, layered structure of vanadium‐based oxides, with the merits of high capacity, structural stability, diverse valence states, and abundant resources, have been reported as cathode materials for ZIBs.^25‐28 However, these cathodes always adopt analogous electrolytes and voltage windows without further investigating their influences on electrochemical properties in aqueous ZIBs.^29,30 In fact, the choice of electrolyte and cut‐off voltage is quite different for vanadium‐based oxide materials as a result of their structural diversity and intercalation chemistry. In previous reports, the influences of different type and concentration of aqueous electrolytes on the performances of Zn‐V₂O₅ cells have been studied by Zhou et al,³¹ demonstrating that the optimization of electrolyte is important for the electrochemical performance in ZIB. Meanwhile, Xu et al³² carefully investigated the effects of cut‐off voltage on the structural stability of NH₄V₄O₁₀ as cathode material in potassium ion batteries. As a result, a de‐ammoniation reaction could be detected at high voltage of 3.9‐4.2 V, leading to poor cycling stability within the voltage of 1 to 4.2 V or 2 to 4.2 V. Controlling the upper charge voltage to 3.8 V, the de‐ammoniation reaction in the charging process is avoided and NH₄V₄O₁₀ electrode displayed stable cycle performance and excellent rate capability, indicating that the structural stability of vanadium‐based oxide materials is sensitive to the cut‐off voltage. In addition to electrolyte and voltage window, crystalline water molecules in the interlayers of vanadium‐based oxides also play a vital role in enlarging the interlayer and supporting the layered structure.^25,33

In this study, we report a layered structure of V₃O₇·H₂O (HVO) cathode with crystalline water molecules inserted into the interlayers through a facile hydrothermal method. The influences of electrolyte and cut‐off voltage on zinc storage behaviors of HVO cathode are investigated, demonstrating the optimized electrolyte and voltage window is 3M Zn(CF₃SO₃)₂ and 0.4 to 1.3 V, respectively. Compared with nonlayered V₃O₇ (VO) nanorod, HVO cathode exhibits high zinc storage capacity and long‐term cycle performance due to the pillar role of crystal water molecules in the layered structure. Furthermore, based on ex‐situ X‐ray diffraction (XRD) and X‐ray photoelectron spectroscopy (XPS) analyses, Zn²⁺ storage mechanism of HVO is recognized as a highly reversible (de)intercalation process with good structural stability.

EXPERIMENTAL SECTION

Material synthesis

Synthesis of HVO nanorods: HVO nanorods were prepared through the hydrothermal method. In detail, 0.364 g V₂O₅ and 0.06 g polyethylene glycol‐400 were dissolved into 25 mL deionized water, while 5 mL H₂O₂ was dropwise added. After stirring for 12 hours at room temperature, the mixed solution was transferred to a 50‐mL Teflon‐lined autoclave and kept at 180°C for 60 hours. The above product was washed with deionized water and ethanol several times and dried at 80°C for 12 hours in a vacuum oven to obtain HVO nanorods.

Synthesis of VO nanorods: HVO nanorods were annealed at 400°C for 2 hours with a heating rate of 1°C per minute under Ar atmosphere. After cooling to room temperature, a VO nanorods were obtained.

Material characterizations

XRD data was carried out at a constant scanning rate of 5° per minute by a Rigaku 2500 X‐ray diffractometer with Cu K_α radiation (λ = 1.5405 Å). The scanning electron microscopy (SEM; FEI Nova Nano‐SEM 230 m) and transmission electron microscopy (TEM; Tecnai G2 F20) were applied to scan the microcosmic crystal morphology and structure. XPS spectra were performed on an ESCALAB 250Xi X‐ray photoelectron spectrometer (Thermo Fisher Scientific).

Electrochemical measurement

The cathode electrode was prepared by coating a slurry of HVO or VO (70 wt%), Super P (20 wt%) and poly (vinylidene fluoride) (10 wt%) on steel foil, and drying at 80°C for 12 hours in a vacuum oven. The working electrode was cut into small pieces with a diameter of 1.2 cm. Then, these electrodes were coupled with zinc foil, electrolytes, and glass fibers separator assemble into CR‐2016 coin cells. Galvanostatic charge/discharge (GCD) curves, cycle performances and rate properties were carried out on LAND CT2001 test system (Wuhan LAND Electronic Co Ltd) at different current densities at room temperature. Cyclic voltammetry (CV) profiles were conducted on the Multi Autolab (M204; Metrohm) at different scan rates and different voltage ranges, in which electrochemical impedance spectrometry (EIS) was also tested within the frequency range from 10⁻² to 10⁵ Hz. Galvanostatic intermittent titration technique was investigated by Arbin instruments.

RESULTS AND DISCUSSION

The working principle of the Zn‐HVO cell is depicted in Figure 1A, in which zinc foil and HVO are employed as anode and cathode material, respectively. As shown in Figure 1B, XRD pattern of HVO demonstrates that all the diffraction peaks are highly indexed to orthorhombic HVO phase (PDF#28‐1433) with the space group of Pnam (unit cell parameters of a = 16.8714 Å, b = 9.3325 Å, and c = 3.6348 Å). The layered HVO, composed of (VO₆) octahedron and (VO₅) triangular pyramid, shows nearly two‐dimensional V₃O₈ layers linked by sharing strong hydrogen bonds along a‐axis and the edge in bc (0kl) planes. To examine the elemental composition of HVO, the XPS spectra of HVO is presented in Figure 1C, in which V 2p_3/2 and V 2p_1/2 peaks at 518.0 and 525.5 eV associated with the oxidation degree of vanadium. Figures 1D and S1 display the TEM and SEM images of HVO, suggesting a typical nanorod morphology with smooth surface. Furthermore, a representative lattice spacing of 0.360 nm is observed in HRTEM image (Figure 1E), which is associated with (230) crystalline face of HVO. Meanwhile, the selected area electron diffraction pattern is presented in Figure 1F, and (130), ($\bar{1}$30), and (060) planes could be detected, indicating an intrinsic single‐crystal structure. In addition, the clear and bright lattice fringes demonstrate that HVO has a high crystallinity degree.

1 Figure. A, Working principle of Zn‐HVO cell; B, XRD pattern and crystal structure of HVO; C, XPS spectra, D, TEM image; E, HRTEM image; F, SAED pattern of HVO. HVO, V3O7·H2O; HRTEM, high‐resolution transmission electron microscopy; SAED, selected area electron diffraction; XPS, X‐ray photoelectron spectroscopy; XRD, X‐ray diffraction

To explore the suitable electrolyte and cut‐off voltage, the electrochemical performances of HVO with different electrolytes and voltage windows are investigated at the current density of 0.5 A g⁻¹. Figure 2A shows the cycling performances of HVO with different electrolytes, in which the cells using Zn(CF₃SO₃)₂ electrolyte present higher capacity than those of ZnSO₄ electrolyte with the same concentration due to the higher reversibility and faster (de)intercalation kinetics of Zn²⁺ ions in Zn(CF₃SO₃)₂ electrolyte.³⁴ In addition, as demonstrated in previous literature,^35,36 high concentrated electrolyte also has a positive effect on the reversible capacity. Therefore, 3M Zn(CF₃SO₃)₂ is chosen as the optimized electrolyte for further study. Zinc storage capacities under different voltage ranges are shown in Figure 2B‐D. In Figure 2B, the charge cut‐off voltage is fixed at 1.4 V and the discharge cut‐off voltage is changed from 0.2 to 0.9 V. When the battery is discharged to 0.2 or 0.3 V, it exhibits higher initial capacity but poor cycling stability, resulting from the unstable crystal structure at excessive discharge state.^16,26 On the contrary, though the battery holds stable cycle performances under discharge voltage from 0.5/0.6/0.7/0.8/0.9 V, its zinc storage capacities are unsatisfactory due to the narrow voltage window.^37,38 Therefore, the discharge cut‐off voltage can be optimized to be 0.4 V, in which the battery could balance the cycling stability and battery capacity. Similarly, 1.3 V is chosen as the upper charge voltage (Figure 2C).³⁹ Thus, 0.4 to 1.3 V is optimized as the ideal voltage window, which can also be verified by the cycling performances in Figure 2D. In addition, the influences of the voltage window on the zinc storage capacity are further systematically investigated through CV curves and GCD profiles. As presented in Figure S2, CV and GCD curves after the first cycle within the voltage window of 0.4 to 1.3 V have a higher coincidence degree, implying reversible cyclic behavior and stable crystal structure.

2 Figure. Cycling performances of the V3O7·H2O electrode at 0.5 A g−1 (A) under different electrolytes, (B) under various discharge cut‐off voltage from 0.2 to 0.9 V (charge cut‐off voltage is 1.4 V), (C) under various charge cut‐off voltage from 1.6 to 0.9 V (discharge cut‐off voltage is 0.4 V), and (D) under different cut‐off voltages

Furthermore, to verify the effects of crystalline water molecules on the structure of HVO, VO is obtained by annealing HVO precursor under an inert atmosphere. The layered‐type structure changes to tunnel‐type after removing crystalline water (Figure S3a). Figure S3b displays the TEM image of VO with a rough surface and nonuniform nanorod structure. It turns out that the crystalline water molecule is critical to the maintenance of the layered structure. Meanwhile, to detect the influences of crystal structure on zinc storage behaviors, CV curves in the initial five cycles of HVO and VO are investigated within 0.4 and 1.3 V. As shown in Figure 3A, two pairs of redox peaks are located around 1.09/0.87 and 0.74/0.51 V, suggesting multistep (de)intercalation mechanism. The differences between the first cycle and subsequent cycles could be attributed to the activation reaction in the initial intercalation process of Zn²⁺.^17,32 Simultaneously, three pairs of redox peaks are observed in the CV curves (Figure S4A) of VO cathode, implying a more complicated (de)intercalation process of Zn²⁺.⁴⁰ Moreover, EIS results presented in Figure S4B and Table S1 further demonstrate that HVO has a smaller charge transfer resistance owing to the favorable layered structure. Figure 3B displays the cycling performances of HVO and VO at the current density of 0.5 A g⁻¹. The specific capacity of HVO is higher than that of VO for 80 cycles, which is also verified by GCD curves in Figure S5. Meanwhile, the coulombic efficiency for HVO is around 100% during cycling, indicating good reversibility of the electrode. There is a gradual increase in capacity at the initial certain cycles, which could be mainly ascribed to the gradual activation process in the layered structure of HVO³³; it is not obvious in VO due to the stable tunnel structure with the insertion of Zn²⁺.^41,42 On the basis of the rate performances in Figure 3C and Figure S6, HVO exhibits reversible capacities of 335, 299, 223, 196, and 178 mAh g⁻¹ at varied current densities of 0.1, 0.3, 0.5, 1, 3, and 5 A g⁻¹, which are higher than those of VO, suggesting excellent rate capability of HVO due to the presence of crystalline water and the stability of layered structure. In addition, the long‐term cycling performances of HVO and VO at a high rate of 5 A g⁻¹ are shown in Figure 3D and the coulombic efficiency is shown in Figure S7. Long‐term cycle measurements further verify that HVO holds good capacity retention of 96% after 2000 cycles, while VO exhibits low discharge capacity but stable cycle performance due to its stable tunneled structure. Moreover, as presented in Figure S8, the XRD pattern of the HVO electrode after 50 and 2000 cycles are almost the same with the initial state, declaring the reversible cycling process and excellent structural reversibility.

3 Figure. A, CV curve of HVO electrode at the scan rate of 0.5 mV s−1 in the initial five cycles. B, Cycling performance of HVO and VO electrode at 0.5 A g−1, C, Rate capability of HVO and VO electrode. D, Long‐term cycling performance of HVO and VO electrode at 5 A g−1. E, CV curves of HVO electrode at different scan rates with the log(current) vs log(scan rate) plots of redox peaks in CV curves. F, Typical capacitive contribution ratio of HVO electrode at 0.8 mV s−1. G, Capacitive contribution of HVO electrode at various scan rates. CV, cyclic voltammetry; HVO, V3O7·H2O; VO, V3O7

To further understand the capacitance contribution in the Zn‐HVO battery, the electrochemical kinetics of Zn²⁺ ions has been studied in detail. CV curves at different scan rates are depicted in Figure 3E. According to the calculation principle Equations (1) to (3) (Supporting Information), the b values of peaks 1 to 4 are calculated to be 0.74, 0.85, 0.80, and 0.84. This implies that the corresponding redox reactions at peak regions are synergistically limited by capacitive and diffusion‐controlled behaviors. Figure 3F,G display the contribution ratios of two capacity mechanism at different scan rates ranging from 0.1 to 1.0 mV s⁻¹. A 58.1% fraction of the total charge at a scan rate of 0.8 mV s⁻¹ is assigned to the capacitive process, accounting for the superior capacitive capability of the HVO electrode. With the scan rate increase, the capacitive contribution ratio of HVO gradually raises from 37.2% to 66.6%. The large capacitive charge storage can be responsible for the high rate capacity.^43,44 The abovementioned pseudo capacitive results indicate that the capacitive effect significantly contributes to the prominent electrochemical behaviors of HVO electrode.^33,45 Figure S9 presents the diffusion coefficients of Zn²⁺ (D_Zn²⁺) in the HVO and VO electrode. As a result, the higher ionic diffusivity of HVO could be attributed to the large interlayer spacing, providing a valid diffusion path for Zn²⁺ and reducing the electrostatic interactions. The comparison of different cathode materials in aqueous ZIBs are shown in Table S2. Compared with other vanadium‐based oxides, the better electrochemical performance and higher capacity retention can be achieved in a narrow voltage range when using HVO as the cathode. Hence, it is important to improve battery performance by optimizing the electrolyte and cut‐off voltage.

Ex‐situ XRD, HRTEM, and XPS are carried out to investigate the zinc storage mechanism and phase evolution of HVO cathode during discharge/charge process. As shown in Figure 4A, it could be clearly distinguished that two peaks at 10.4° and 32.5° (match with the crystal face of (020) and (330), respectively) gradually shift to higher 2θ degrees in the discharge process, representing the decrease of the interplanar space, that is, the a and b lattice paraments. Along with the increasing depth of discharge, raising x in Zn_xV₃O₇·H₂O leads to the location of characteristic peaks shift to the lower diffraction angle. With the increasing content of Zn²⁺, a small contraction of interlayer space is observed during the discharge process because the inserted Zn²⁺ with crystalline water can shield the layer electrostatic repulsion. Moreover, the formation of hydrogen bonds among Zn²⁺, H₂O, and lattice oxygen bring the bilayer closer together. Changes in the interlayer space at full discharge/charge states are illustrated in HRTEM images (Figure 4B,C). The (330) peak in XRD shifts to 33.2° and the relevant interlamellar spacing shrinks from 0.273 to 0.269 nm when discharged to 0.4 V. After charging to 1.3 V, the peak could return to original location and the lattice spacing of (330) plane returns to 0.273 nm, indicating that the layered structured HVO has a good structural reversibility.

4 Figure. A, Ex‐situ XRD patterns and corresponding galvanostatic charge‐discharge curve at the current density of 0.2 A g−1; HRTEM image in fully (B) discharged and (C) charged states of HVO electrode; XPS spectra of (D) Zn 2p and (E) V 2p in the initial, fully discharged and charged states. F, Schematic illustration of zinc storage mechanism in HVO. HVO, V3O7·H2O; XPS, X‐ray photoelectron spectroscopy; XRD, X‐ray diffraction

XPS spectra of HVO at fully discharged and charged states are carried out to confirm the changes in valence state of elements. As presented in Figure 4D, Zn 2p spectra are strong evidence to disclosing the reversible (de)intercalation mechanism of Zn²⁺ in HVO electrode. At the initial state, there is no Zn 2p signal detected, while two characteristic peaks appear at 1021.9 and 1044.8 eV (corresponding to Zn 2p_3/2 and 2p_1/2, respectively) after discharged to 0.4 V, demonstrating the insertion of Zn²⁺ ions. Furthermore, after being charged to 1.3 V, Zn 2p peaks could be still discovered but with lower intensity than that of the discharged state, verifying the residue of Zn²⁺ ions in HVO. On the basis of the XPS spectrum of V 2p in Figure 4E, the valence state of V⁵⁺ is totally transformed into V⁴⁺ due to the insertion of Zn²⁺ ions. In addition, the signal area of V⁵⁺ is approximately 1.5 times larger than that of V⁴⁺ in the initial state, and the value is changed to 1.2, indicating the increase of V⁴⁺ and the residue of Zn²⁺ ions in the fully charged state. The energy storage mechanism of the HVO cathode is displayed in Figure 4F. The active inserting sites of Zn²⁺ ions in the discharge process are mainly located at the space between layers of HVO, and EDS result in Figure S10 indicates the existence of Zn²⁺. In the charge process, most of the inserted Zn²⁺ ions could reversibly extract from the layers and maintain structural stability of HVO (Figure S11).

CONCLUSION

In conclusion, the layered HVO cathode is successfully prepared by a facile hydrothermal method, and the electrolyte and cut‐off voltage are further investigated. On the basis of the optimized electrolyte (3M Zn(CF₃SO₃)₂) and cut‐off voltage (0.4‐1.3 V), HVO delivers high capacity and excellent cycle stability under high current density. Compared with the nonlayered structure cathode (V₃O₇), it is confirmed that the presence of crystal water is a key factor for HVO to maintain layered structure. As a result, HVO presents superior electrochemical properties and faster Zn²⁺ diffusion kinetics. Furthermore, the Zn²⁺ storage mechanism of HVO could be identified as a reversible insertion/extraction reaction with high structural stability, implying its potential application in the field of large‐scale energy storage.

ACKNOWLEDGMENT

This study was supported by the National Natural Science Foundation of China (Grant no. 51932011, 51972346, 51802356, and 51872334) and Innovation‐Driven Project of Central South University (No. 2020CX024).

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.