The V₂O₅·nH₂O nanowires were prepared through a hydrothermal method. CNTs were incorporated to increase the conductivity. X-ray diffraction (XRD) patterns in Figure 1(a) show the crystal structures of CNTs, V₂O₅·nH₂O, and V₂O₅·nH₂O/CNT. CNTs have two broad peaks centered at 26.1° and 43.0°, attributed to the (002) and (100) planes of graphitic carbon, respectively. The prominent peaks for V₂O₅·nH₂O and V₂O₅·nH₂O/CNT locate at 15.4°, 20.3°, 21.7°, and 31.1°, which are assigned to (200), (010), (110), and (310)/(400) planes of orthorhombic V₂O₅ (PDF#72-0598) as marked in the figure [23]. The thermogravimetric and differential thermal analysis (TGA-DTA) profiles of neat V₂O₅·nH₂O (Figure S1) confirm the presence of 1.4 wt.% water, corresponding to a chemical composition of V₂O₅·nH₂O (n\approx 0.14). The surface areas of V₂O₅·nH₂O and V₂O₅·nH₂O/CNT are 13.5 and 43.6 m² g^-1, respectively (Figure S2). The high surface area of the latter mainly comes from the incorporation of 33.7 wt.% CNTs. The scanning electron microscopy (SEM) and optical images of the V₂O₅·nH₂O/CNT film are exhibited in Figure 1(b), revealing a flexible and porous structure that consists of intertwined CNTs and V₂O₅·nH₂O nanowires. The nanowire has a diameter of around 100 nm with high crystallinity, as shown in the transmission electron microscopy (TEM) images (Figures 1(c) and 1(d)). The crystal lattice in Figure 1(d) gives a spacing of 0.34 nm, consistent with the (101) plane of V₂O₅·nH₂O. The selected area electron diffraction (SAED) in the inset further confirms the orthorhombic structure of V₂O₅·nH₂O. The films are directly used as electrodes to evaluate their Zn ion storage behavior in different electrolytes.

To explore the electrolyte effect, the discharge/charge profiles and cyclic performances of flexible V₂O₅·nH₂O/CNT electrodes tested in 1 M Zn-H₂O and 1 M Zn-EC/EMC electrolytes (see Experimental) are compared at a low current rate of 100 mA g^-1 (Figures 2(a) and 2(b)). The voltage profiles in the two electrolytes have similar shapes whose sloping features suggest the solid-solution reaction processes associated with Zn ion intercalation/deintercalation [12, 24]. It is noted the irreversible charge capacity in the first cycle for the ZIB in Zn-H₂O may be due to the intrinsic dissolution of V₂O₅·nH₂O in H₂O, similar to the shuttle effect in lithium-sulfur batteries [25]. For the cyclic test, it shows a fast capacity degradation of the cell in aqueous electrolyte. The capacity decreases from 435 mAh g^-1 in the first cycle to 70 mAh g^-1 after 40 cycles. In contrast, the capacity of the cell in Zn-EC/EMC slightly decreases in the initial cycles and stabilizes at 382 mAh g^-1 after 40 cycles, demonstrating a much more stable cyclic stability.

The crystal structure evolution was probed by ex situ XRD at different charge/discharge depths to explore the reasons for the stability in organic electrolyte, as shown in Figure 2(c). There is no obvious change of the phase after discharging to 0.9 V. Several new peaks appear starting from 0.5 V, and the intensity increases with further discharging to 0.2 V. These peaks located at around 9.2°, 12.2°, 33.2°, and 35.9° correspond to a new phase, similar to previous report [8, 24], but their detailed composition is hard to be determined at this stage. Two peaks at 15.4° and 31.0° present downshifts to 14.1° and 28.6°, respectively. The merging of two peaks (20.3° and 21.8°) into one at 20.5° and the presence of a minor peak at 18.8° are due to the increased Zn ion content intercalated into V₂O₅·nH₂O [16, 20]. Upon charging, the shifted peaks recover to 15.4° and 31.0° at 1.0 V and the additional peaks disappear with a continuous charge to 1.6 V, indicating the reversibility of phase transitions. The structural change in aqueous electrolyte shows a similar trend with that in the organic electrolyte (Figure S3). However, the diffraction peaks in the cycled electrode in Zn-EC/EMC have much higher intensities than those in Zn-H₂O, implying better crystalline degree after the full charge in organic electrolyte. This phenomenon becomes more apparent with increasing cycle numbers. The XRD patterns of the electrode after 100 cycles in aqueous electrolyte fully lose the crystallinity, compared to intact crystal structure for the electrode cycled in organic electrolyte (Figure 2(d)). It is noted the electrode in Zn-EC/EMC maintains its freestanding property after 100 cycles and could be peeled off from the titanium current collector for XRD test, whereas the one cycled in Zn-H₂O shows the collapsed structure and closely attaches to the titanium foil. Therefore, Ti peaks are observed in the XRD pattern of Zn-H₂O electrode. To examine whether the organic electrolyte affects the charge storage mechanism, X-ray photoelectron spectroscopy (XPS) measurement was further conducted for the electrodes cycled in Zn-EC/EMC. Figure 2(e) shows the XPS deconvoluted V 2p spectra of electrodes in pristine and fully discharged/charged states. Two peaks located at 517.6 and 525.0 eV for the pristine V₂O₅·nH₂O are indicative of the V⁵⁺ signal. When discharged to 0.2 V, additional peaks centered at 524.2/516.8 eV and 523.5/516.1 eV are assigned to V⁴⁺ and V³⁺, respectively. The appearance of these peaks implies the reduction of V⁵⁺ due to Zn ion intercalation into V₂O₅·nH₂O. The peaks are recovered upon charging as a reflection of the deintercalation of Zn ions. Combined with XRD and XPS results, it is suggested that similar electrochemical process occurs in the aqueous and organic electrolyte, both of which show excellent reversibility of Zn ion intercalation/deintercalation.

To figure out the underlying mechanisms, the morphologies of V₂O₅·nH₂O/CNT electrodes after cycles were examined by TEM and SEM. The V₂O₅·nH₂O after the 1st full discharge in Zn-EC/EMC shows the stable nanowire morphology with the lattice structure changed to be less ordered as shown in Figures 3(a) and 3(b), arising from the insertion of Zn ion into the layers of V₂O₅·nH₂O. The Zn element is homogeneously dispersed in the V₂O₅·nH₂O wire as reflected by the scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) image in the inset of Figure 3(a). After the 1st charge, the crystal structure of V₂O₅·nH₂O is recovered (Figures 3(c) and 3(d)), indicative of the excellent structure reversibility of electrochemical reactions in Zn-EC/EMC. In contrast, the nanowires in Zn-H₂O after both 1st discharge and charge display a loose and porous morphology with a disordered structure, as shown in Figures 3(e)–3(h) and S4. Although the SAED pattern of the electrode after 1st charge in Zn-H₂O presents the crystal spots, the irreversible structure damage has already occurred. While the inner part of the fiber maintains the crystal structure, the outer section has already amorphized as observed in Figures S5a-5c. The continuous amorphization with repeated cycling leads the completed collapse of the fiber structure, which is in agreement with the XRD results (Figure S5d). Therefore, the morphologies of V₂O₅·nH₂O/CNT electrodes after 100 cycles in Zn-H₂O and Zn-EC/EMC were examined to further investigate the stability as displayed in Figures 3(i) and 3(j). It is noted that the electrode in Zn-EC/EMC maintains the nanowire structure, whereas the one cycled in Zn-H₂O almost disappears. The energy-disperse spectroscopy (EDS) mapping results in Figure 3(k) indicate the one-dimensional structure of V₂O₅·nH₂O in Zn-H₂O is completely collapsed and the active materials are merged with the CNT network.

It is speculated that the damage of the electrode structure is related to the dissolution of active materials. The solubility of V₂O₅ in water is 0.8 g L^-1 at 20°C [26]. In an effort to identify the stability of V₂O₅·nH₂O/CNT electrodes in electrolytes, the electrode materials (V₂O₅·nH₂O) with identical mass (0.7 mg) were separately added in 1 mL electrolytes of Zn-EC/EMC, Zn-H₂O, and their hybrids with water to EC/EMC in a volume ratio of 0.5 : 9.5, 1 : 9, 2 : 8, 3 : 7, 4 : 6, and 5 : 5. The color of Zn-EC/EMC remained transparent while the color of Zn-H₂O changed to light yellow after 3 days, as shown in the inset of Figure 4(a). To study the relationship of the solubility and the electrolyte, inductively coupled plasma mass spectrometry (ICP-MS) was conducted to measure the concentrations of vanadium element in different electrolytes after 3 days. As shown in Figure 4(a), the content of vanadium element shows a slight increase from 0.49 mg L^-1 in Zn-EC/EMC to 0.84 mg L^-1 in Zn-H₂O-EC/EMC(2-8) but surges to 2.47 mg L^-1 in Zn-H₂O-EC/EMC(3-7) and further to 46.7 mg L^-1 in Zn-H₂O. For the Zn-EC/EMC, the concentration of vanadium is only 0.63 mg L^-1 even after 4 months, demonstrating the low solubility of V₂O₅·nH₂O in the organic electrolyte. The nonlinear solubility with respect to H₂O content implies the solvent-solvent interaction and the solute-solvent interaction in the electrolytes [27]. The above finding signifies that the electrode in Zn-EC/EMC shows much better structural stability than that in Zn-H₂O, so as to largely improve the capacity retention. In addition to the solubility, the electrolyte also affects the charge transfer. The electrochemical impedance spectroscopy (EIS) was conducted to examine the internal resistance of the cells in both two electrolytes. The characteristic resistance values calculated according to the equivalent circuit are shown in the inset of Figure 4(b) [28, 29]. R_{\text{e}} is the resistance of electrolyte; Rst//CPEst represents the interphase contact resistance; Rct//CPEdl is the charge transfer resistance, and W_0 refers to the diffusion resistance. The fresh cell in Zn-H₂O presents much lower values of R_{\text{ct}} and R_{\text{st}} than the cell in Zn-EC/EMC, due to the better wettability of both anodes and cathodes and the lower desolvation energy of Zn ions in aqueous electrolytes [17]. To reduce the charge transfer resistance and maintain the stability, an aqueous/organic hybrid electrolyte is therefore designed with various ratios. We find that with 10% H₂O in EC/EMC as solvent, both interfacial and charge transfer resistances are largely reduced from 580.6 and 561.3  Ω to 1.6 and 298.0  Ω, respectively, while the dissolution of V₂O₅·nH₂O is minimized as evidenced by low solubility of vanadium in Figure 4(a). Further increasing the water content is helpful in improving the kinetics, but leads to deterioration of the electrode. The stability of flexible V₂O₅·nH₂O/CNT electrodes in different electrolytes is then evaluated. The electrodes were firstly activated at a low current rate of 50 mA g^-1 for two cycles, as shown in Figure 4(c), before charge/discharged at 500 mA g^-1. The cells have fast capacity degradations using the electrolytes of Zn-H₂O and Zn-H₂O-EC/EMC(4-6). The residual capacities are only 84 and 302 mAh g^-1 after 100 cycles from initial capacities of 394 and 465 mAh g^-1, respectively. In contrast, for Zn-H₂O-EC/EMC(1-9), the capacities of cells rise gradually in the initial cycles for material activation and stabilize at 549 mAh g^-1 after 100 cycles, which is even higher than its initial capacity. As to Zn-EC/EMC, the cell delivers a continuously increased capacity along with cycles. The increase in the capacity is ascribed to the decreased resistance after cycling (Figure S7, EIS), probably attributed to the close interphase contacts among anode, cathode, and electrolyte and the electrochemical polishing of passive film on the surface of Zn anode [3, 17]. The morphology of the electrode after 100 cycles in Zn-H₂O-EC/EMC(1-9) was also examined to further confirm the excellent structural stability of V₂O₅·nH₂O in the hybrid electrolyte, as shown in Figure S6a. More water contents in the electrolyte lead to in the loose and collapsed structures after cycles (Figures S6b-S6e), which is consistent with the solubility result (Figure 4(a)).

The comparison of rate capabilities in different electrolytes measured at increasing current rates from 0.1 to 0.2, 0.5, 1, and 2 A g^-1 is given in Figure 4(d). The cell in Zn-H₂O delivers fast capacity degradations at all the current rates. Even when the rate is reduced back to 0.1 A g^-1, the capacities still continuously decrease. Under the Zn-EC/EMC electrolyte, the V₂O₅·nH₂O/CNT electrode delivers capacities of 301, 268, 210, 169, and 129 mAh g^-1, respectively. When the rate is reduced back to 0.5 and 0.1 A g^-1, the capacities are recovered to 256 and 371 mAh g^-1. The hybrid electrolyte is expected to achieve a synergistic effect in obtaining both the structural stability and superior rate capability. The organic solvent effectively enhances the stability of the electrode, and the incorporated water in the hybrid electrolyte improves the ionic conductivities as discussed before. Therefore, the electrodes in Zn-H₂O-EC/EMC(4-6) and Zn-H₂O-EC/EMC(1-9) show attractive capacities under all the current rates. The latter has higher value due to its advantage in inhibiting the dissolution of active materials (Figure 4(a)). The long-term cyclic test of the cell was therefore conducted in Zn-H₂O-EC/EMC(1-9) at a high current density of 4 A g^-1 as shown in Figure 4(e). The battery takes 50 cycles at 0.5 A g^-1 to activate the materials and maintains a high capacity of 282 mAh g^-1 after 1000 cycles. The efficiencies constantly sustain around 100% during the cyclic process. A comparison of the electrochemical performance between the current work and the vanadium-based electrodes is presented in Table S1. Among these studies, the current V₂O₅·nH₂O/CNT electrodes deliver equally excellent or even better capacities for long cycles due to the benefits of hybrid electrolytes.

Fast self-discharge is a general issue facing the aqueous batteries [30]. To explore the self-discharge rate with hybrid electrolyte, the performance of these cells is compared in Figure 5. All the batteries were cycled at 500 mA g^-1 for 40 cycles to activate the electrode. They were then rested for 72 h before further discharging. The battery in Zn-H₂O only releases a 35.1% capacity of the charged value, corresponding to a fast self-discharge rate of 0.9% per hour. This is not surprised considering the fast dissolution of V₂O₅·nH₂O in aqueous electrolyte. Along with time, active materials will be dissolved and shuttled to the anode, giving rise to a rapid self-discharge. Turning to the organic and hybrid electrolytes (10% H₂O), the cells deliver identical discharge capacities to the charged ones. The dissolution phenomenon is totally inhibited in the two electrolytes for avoiding fast self-discharge.

The formation of zinc dendrites on the anode during cycling is always a severe issue to influence the electrochemical performance of cells [31–33]. Continuous growth of dendrites may penetrate the separator, leading to the short circuit of batteries. This phenomenon is effectively alleviated in hybrid electrolyte. The cycled cells were disassembled as shown in Figure S8. The Zn anode is severely eroded in the aqueous electrolyte in contrast to the intact electrode in organic and hybrid electrolyte. The morphologies of Zn electrodes before and after 100 cycles were further examined by SEM. Figure 6(a) presents the pristine Zn foil with a very smooth surface. After cycling in Zn-H₂O, it shows a rough surface covered by a number of Zn rods, compared to the flat texture for the Zn tested in Zn-H₂O-EC/EMC(1-9) and Zn-EC/EMC (Figures 6(b), 6(c), and 6(d)). To explore the individual effect of EC and EMC, symmetrical cells were cycled using two identical Zn foils and electrolytes of Zn-H₂O, Zn-H₂O-EC, Zn-EC, Zn-H₂O-EC/EMC(4-6), Zn-H₂O-EC/EMC(1-9), and Zn-EC/EMC. It is noted that EMC and H₂O are immiscible and Zn(ClO₄)₂ is insoluble in EMC (Figure S9). EC plays twofold effects. On the one hand, it increases the solubility of the salt in organic solvent thanks to its large dielectric constant. On the other hand, it facilities the miscibility of H₂O, EC, and EMC mixture enabling the preparation of hybrid electrolyte [34, 35]. All the cells were tested at a current density of 2 mA cm^-2 with a Zn plating capacity limitation of 2 mAh cm^-2. After 50 cycles, Zn electrodes were examined by SEM, as shown in Figure S10. It shows severe Zn dendrites appear on the surfaces in Zn-H₂O, Zn-H₂O-EC, and Zn-EC, especially for Zn-EC where dendrites are most pronounced. However, with the presence of EMC, Zn dendrites are effectively alleviated. The suppression of dendrite growth is enhanced with increasing EMC contents, as presented in Figures S10d-S10f. The mechanism may reside in the blocking effect of alkyl groups on EMC that attaching and covering the surface of Zn electrodes, specifically the active sites where Zn dendrite grows rapidly [36].

To study the stability of Zn anodes during cycles, symmetrical cells with electrolytes of Zn-H₂O, Zn-H₂O-EC/EMC(1-9), and Zn-EC/EMC were electrochemically measured. Figure 6(e) presents a comparison of plating and stripping of Zn//Zn electrodes in different electrolytes at a current density of 2 mA cm^-2 with a capacity limitation of 2 mAh cm^-2. In Zn-H₂O, low overpotentials of about 59 mV are obtained in the first several cycles. However, some fluctuations with abrupt augments are observed arising possibly from the nonuniform plating/stripping. As to the cells with Zn-H₂O-EC/EMC(1-9) and Zn-EC/EMC, although the overpotentials for the first cycle are as large as 216 and 217 mV, respectively, they gradually decrease with prolonging cycle numbers (Figures 6(f) and S11), as a result of the reduced internal resistance of batteries after cycles [17]. The above observations suggest the beneficial effect of hybrid electrolyte in stabilizing not only the cathode but also Zn anode.

V₂O₅·nH₂O nanowires were synthesized via the hydrothermal method [23]. 0.3 g of commercial V₂O₅ powders (≥98%, Aldrich) was added in 30 mL deionized (DI) water and vigorously stirred for 10 min. 2 mL of 30 wt.% H₂O₂ solution was dropwisely added to the V₂O₅ dispersion, followed by stirring until the dispersion changed to a red-brown sol. The sol was then transferred into a 50 mL Teflon-lined autoclave and treated by a hydrothermal process at 220°C for 48 h. The resulted yellow powders were washed with DI water and dried at 60°C for 12 h to obtain V₂O₅·nH₂O nanowires. For preparing V₂O₅·nH₂O/CNT films, 20 mg V₂O₅·nH₂O and 10 mg CNTs (Timesnano, Chengdu) were mixed in 500 mL DI water by sonication for 30 min to form a uniform dark-green dispersion. The films were obtained by vacuum filtration of the dispersions through a membrane filter (Millipore, 220 nm pore size, 35 mm in diameter), which were dried at 60°C for overnight before peeling off from the filter paper.

The morphologies of V₂O₅·nH₂O/CNT films before and after cycles were examined on scanning electron microscopes (SEM, 6335F and TESCAN VEGA3) and a transmission electron microscope (TEM, JEOL 2100F). Spectrum imaging of electron energy loss spectroscopy (EELS) was carried out under 200 kV accelerating voltage with a 13 mrad convergence angle for the optimal probe condition. Energy dispersion of 0.7 eV per channel and 21 mrad collection angle were set up for EELS; HAADF images were acquired with an 89 mrad inner angle simultaneously. The crystalline phases of the materials were detected on an X-ray diffraction (XRD) system (Rigaku SmartLab) with Cu Ka radiation source. To determine the water and CNT content in V₂O₅·nH₂O/CNT, thermogravimetric and differential thermal analysis (TGA-DTA) was conducted in the temperature range of 25-600°C at a heating rate of 10°C min^-1 under N₂ and air atmosphere, respectively. X-ray photoelectron spectroscopy (XPS, PHI5600 by Physical Electronics, Inc.) was conducted using monochromatic Al Ka X-ray at 14 kV. The solubilities of V₂O₅·nH₂O in different electrolytes were measured by inductively coupled plasma mass spectrometry (ICP-MS).

The V₂O₅·nH₂O/CNT films were cut into electrode discs of 12 mm in diameter, ~1.6 mg cm^-2 of active materials in weight, and ~70  μm in thickness. CR2032 coin cells were assembled in the ambient using the Zn foil as the anode and the glass fiber (Whatman, GF/D) as the separator. A titanium foil was adopted as the current collector to avoid the corrosion. To investigate the solvent effect, 1 M zinc perchlorate (Zn(ClO₄)₂) in organic solvents (ethylene carbonate (EC) and ethyl methyl carbonate (EMC) at a volume ratio of 1:1), DI water and water/organic hybrids were used as electrolyte. The coin cells with different electrolytes were charge/discharge cycled between 0.2 and 1.6 V on a LAND 2001 CT battery tester at room temperature. The electrochemical impedance spectra (EIS) were obtained at a constant perturbation amplitude of 5 mV in the frequency range between 0.01 Hz and 100 kHz on a VMP electrochemical workstation (Biologic, France).