Background

Vanadium oxides (e.g., V6O13, V3O7, V2O5) are cathode materials applicable for high-energy lithium-ion battery (LIB), due to their low cost, high specific capacities, and the abundance of vanadium element [1,2,3,4,5,6]. Among the oxides, V6O13 has been considered as an excellent candidate of the cathode material [7,8,9,10,11,12,13,14]. Its theoretical capacity and energy density can reach 417 mA h g−1 and 890 Wh kg−1 when lithiated to the final product, Li8V6O13 [2, 8]. However, V6O13 electrodes have suffered short cycle life and low rate capability for a long time because V6O13’s electronic conductivity decreases when lithiated while Li+ diffusion coefficients (10−8 to 10−9 cm2 S−1) are low [7, 9]. Constructing free-standing 3D nanostructures is an effective method to solve the above problems. 3D nanostructure can enhance ion/electron transport/diffusion while effectively avoids self-aggregation [15,16,17,18,19,20]. For example, Yu et al. synthesized 3D V6O13 nanotextiles assembled from interconnected nanogrooves via a facile solution-redox-based self-assembly route with MnO2 template at room temperature. In a voltage range of 1–4 V, V6O13 nanotextiles exhibited reversible capacities of 326 and 134 mA h g−1 at 20 and 500 mA g−1, respectively, and a capacity retention of above 80% after 100 cycles at 500 mA g−1 [2]. Tong et al. fabricated V6O13 cathode supported by a steel mesh with wrinkles by the similar route. The free-standing electrode with a loading amount of V6O13 up to 2.0 mg cm−2 was obtained. At a current density of 500 mA g−1, the V6O13 electrode demonstrated an initial capacity of 225 mA h g−1 that deteriorated to around 150 mA h g−1 after 500 cycles [21]. However, above research involved the two-step electrodeposition and removal of MnO2. Direct growth of mixed-valence vanadium oxide nanostructure with good electrochemical property remains a great challenge [22]. Meanwhile, previous studies have not demonstrated V6O13-based flexible cathode, which has a potential use in the wearable devices.

Herein, we proposed a simple hydrothermal process to successfully grow interconnected V6O13 nanosheets on the carbonized textile to fabricate a 3D free-standing electrode. It exhibited specific capacities of 161 and 105 mA h g−1 at the specific currents of 300 and 1200 mA h g−1, respectively. With carbon nanotube (CNT) coating to further improve its conductivity, its specific capacities increased to 170 and 140 mA h g−1. Meanwhile, its cycling performance was also improved. It could retain 74% of initial capacity with CNT coating, compared with 50% retention without CNT coating after 400 cycles at 300 mA g−1. The improvement on the electrochemical performance was mainly ascribed to the synergistic effect of the 3D nanostructure of V6O13 and hierarchical conductive network.

Methods

Synthesis of c-textile

The commercially available bamboo cloth was soaked in a solution with 2.5 g NaF and 60 ml H2O for 1 h and dried for 5 h in 120 °C oven. The dried textile was carbonized at 800 °C in N2 for 30 min to obtain c-textile.

Growth of 3D V6O13 Nanostructure on c-textile

3D V6O13 nanostructure was grown on c-textile by a seed-assisted hydrothermal method. c-textile was slightly oxidized in the condensed nitric acid (80 wt%) for 30 min. V2O5 powder (1 mg) was added to 5 ml deionized water and then ultrasonicated for 15 min to obtain a suspension. The oxidized c-textile was then immersed into the suspension for 2 h, dried, and heated at 300 °C for 10 min to grow the seed of vanadium oxide on c-textile. V2O5 powder (16 mg) was added to 224 μl of 30 wt% H2O2 and stirred for 10 min to obtain a brown solution. It was then diluted with additional 40 ml distilled water and stirred for 30 min. After the solution was transferred into a 25-ml stainless-steel autoclave, the oxidized c-textile was immersed into the solution. The autoclave was kept at 180 °C for 48 h, then the sample was washed with distilled water and alcohol and dried at 60 °C for 8 h to finally get the flexible 3D free-standing V6O13 nanostructure supported with flexible c-textile. CNT was further coated on V6O13 nanostructure by repeatedly dipping it into NMP suspension (0.5 mg/mL) of multi-walled CNT and drying to produce a V6O13/CNT composite electrode.

Characterization of Materials

Morphology of the product was observed by a scanning electron microscopy (SEM, Philips XL30 FEG) and a transmission electron microscopy (TEM, JEOL JEM-2010). X-ray photoelectron spectroscopy (XPS) analyses (K-Alpha) were performed using a monochromatic Al Ka source.

Battery Fabrication and Electrochemical Measurements

Standard CR2016-type coin cells were assembled in an argon-filled glove box (Vigor Inc. Suzhou, China) with V6O13 electrode as the working electrode with a mass loading of ~ 1 mg cm−2. A lithium foil was used as the counter electrode; 1 mol LiPF6 in a mixture of ethylene carbonate (EC), diethyl carbonate (DEC), and dimethyl carbonate (DMC) with a volume ratio of 1:1:1 was used as the electrolyte, and a polypropylene film was used as the separator. The assembled cells were electrochemically cycled between 1.5 and 4.0 V vs. Li/Li+ for galvanostatic charge/discharge on a LAND battery test system (Wuhan Kingnuo Electronics Co., Ltd., China) at 25 °C. Electrochemical impedance spectroscopy (EIS) studies were conducted with Autolab PGSTAT302N workstation in the frequency range of 10 mHz to 10 kHz.

Results and Discussion

The schematic of the growth of 3D V6O13 interconnected nanosheets on c-textile was shown in Additional file 1: Figure S1. The textile (Fig. 1a) was firstly carbonized at 800 °C to obtain c-textile (Fig. 1b). SEM images (Fig. 2a) showed that c-textile was composed of weaved bundles of carbon fibers with a diameter of ~ 5 μm. c-textile exhibited excellent flexibility and mechanical strength. It was able to be rolled and twisted as shown in Fig. 1c. The square resistivity of the c-textile was measured to be 5 Ω/sq. with the four-probe method. Thus, it was used as a promising flexible support/collector for the electrode materials. It was then immersed in VOx suspension, dried, and kept at 300 °C for 10 min to grow the seed crystals. Its weight change was undetectable (< 0.1 mg). After being immersed in vanadium oxide (VOx) sol solution for the hydrothermal growth, the black c-textile was covered with a layer of a yellow-green thin film; however, its flexibility was kept, as shown in Fig. 1d. Its resistivity increased to 50 Ω/sq. SEM images (Fig. 2b, c) further showed that it was composed of several micron-long and several hundred nanometer-wide interconnected nanosheets, as building blocks to construct 3D nanostructure on c-textile. High-resolution TEM image (Fig. 2f) showed well-defined lattice fringes of the grown nanosheets. The spacing of 3.5 Å in the lattice fringe was consistent with (110) interplanar distance of the orthogonal V6O13 phase (PDF card No.71-2235) which was in agreement with the XRD pattern (Fig. 3c). The growth mechanism was that seed crystal firstly nucleated on the sites with an oxygen-bearing functional group [23, 24]. Then during the hydrothermal process in the VOx aqueous solution, interconnected V6O13 nanosheets were continuously grown on the seed crystals. As for the formation of 3D structured microflowers, it might be due to the several seed crystals aggregated at the same location for growth of the nanosheets. To further determine the valence state of vanadium element in V6O13, XPS analyses were conducted on the synthesized interconnected V6O13 nanosheets. The survey XPS scan (Fig. 3a) revealed that the sample was composed of V, O, C, and N elements. The binding energies for vanadium 2p3/2 and 2p1/2 were identified in Fig. 3b at 516.0 and 523.9 eV for V4+ and 517.3 and 525.0 eV for V5+, respectively. It was well consistent with the chemical state of vanadium in V6O13 reported [25,26,27]. The above results confirmed that 3D V6O13 nanostructures were successfully grown on c-textile via a simple seed-assisted hydrothermal process.

Fig. 1
figure 1

Optical images of a commercially available textile, b carbonized textile, c rolled and twisted c-textile, and d c-textile with grown V6O13 at the rolled state, inset: at the flat state

Full size image
Fig. 2
figure 2

Microstructure of 3D free-standing interconnected V6O13 nanosheets on c-textile: a, b low resolution SEM images of c-textile without and with nanosheets, respectively; c, d high-resolution SEM images of interconnected nanosheets grown on c-textile; e, f low- and high-resolution TEM images of the nanosheet, respectively

Full size image
Fig. 3
figure 3

a A survey XPS spectrum of 3D free-standing interconnected V6O13 nanosheets grown on c-textile. b Spectrum of V2p and O1s with a fitted curve. c The XRD patterns of V6O13

Full size image

To investigate the electrochemical performance of 3D V6O13 nanostructures grown on c-textile, half-cell coin batteries were assembled with a V6O13 electrode as the cathode and lithium foil as the anode. Figure 4a displayed typical cyclic voltammetry (CV) curves of V6O13 electrode in a scan rate of 0.2 mV s−1 between 1.5 and 4.0 V (vs. Li/Li+). The main redox peaks at 2.8/2.5 V could be easily identified. Broad anodic peak at ~ 3.2 and 2.3 V and cathodic peak at ~ 1.8 V could also be observed [11, 28]. The appearance of the above peaks indicated multi-step phase transitions, and the location of peak voltages was consistent with the previously reported ones [2]. Figure 4b showed the galvanostatic discharge/charge curve for the first cycle at the specific current of 30 mA g−1. Multiple poorly defined voltage plateaus could be identified. In the discharge curve, two sloped plateaus were identified at 2.3 and 2.8 V, corresponding to the anodic peaks. When the specific currents increased from 30 to 150, 300, 600, and 1200 mA g−1, the specific capacities were 253, 176, 161, 133, and 105 mA h g−1. The good electrochemical properties were due to the 3D nanostructure composed of V6O13 nanosheets. Such open structure could contact with electrolytes very well and shorten the Li+ transport and diffusion path. The morphology of the original V6O13 electrode and the cycled electrode with SEI was examined under SEM (Additional file 1: Figure S2). The morphology of 3D interconnected nanosheets was retained during the cycling. This further suggests the importance of the 3D nanostructure on the structural integrity of V6O13 electrode during the electrochemical cycling. However, the conductivity of V6O13 decreased as lithiation proceeded [7, 29]. Moreover, the length of nanosheet in the V6O13 electrode reached several tens of microns. Only a small portion of individual nanosheet is directly connected with the conductive carbon fibers of c-textile, which served as the collector. It could hinder the electron transfer during the charge/discharge process. To further enhance the conductivity, and thus the electrochemical properties of the 3D free-standing V6O13 electrode, it was immersed in CNT dispersion to dip-coat CNT on its surface. Figure 5a, b showed SEM images of the V6O13 electrode with CNT. CNT was successfully deposited in the plane of V6O13 nanosheets and intimately contacted with them. Even the bridging was built through CNTs between neighboring nanosheets, as shown in Fig. 5c. As expected, the resistivity of the V6O13 electrode with CNT decreased from 50 to 20 Ω/sq. After the coating of CNTs, the redox peak appeared at the same position on CV profile (Fig. 4a) while peak currents increased. It indicated fast kinetic of electrochemical reaction in V6O13 electrode with CNT. The V6O13 electrode with CNT exhibited better rate performance compared with the electrode without CNT coating, as shown in Fig. 4c. The specific discharge capacities were 261, 185, 170, 153, and 140 mA h g−1 at the specific currents of 30, 150, 300, 600, and 1200 mA g−1, respectively, corresponding to 12~40% increase compared with the composite cathode without CNTs. To further verify the role of CNT, we calculated the lithium ion diffusion coefficient with cyclic voltammetry. The V6O13/CNT anodic and cathodic diffusion coefficients were 4.79 × 10−8 and 2.01 × 10−8 cm2 s−1, higher than V6O13 electrode’s 2.42 × 10−8 and 1.7 × 10−8 cm2 s−1, respectively (and the associated discussion is in Additional file 1: Figure S3). Nyquist plots (Fig. 6a) of V6O13 electrode and V6O13 electrode with CNT displayed similar shapes, a semicircle shape in the high-to-medium frequency domain and an inclined line in the low-frequency regions, corresponding to electrochemical reaction impedance (charge transfer process) and diffusion process of lithium ions. The inset is the equivalent circuit used to fit Nyquist plots. In the circuit, CPE is the constant phase angle element and W is the Warburg impedance. Rs and Rct represent the ohmic resistance (total resistance of the electrolyte, separator, and electrical contacts) and the charge transfer resistance, respectively [22, 30]. Additional file 1: Table S1 listed the parameters used to fit the plots. Rct for the V6O13/CNT electrode was calculated to be 37.24 Ω, lower than that of V6O13 (55.58 Ω). This decrease in charge transfer resistance was ascribed to the addition of CNT. The mechanism was illustrated in Fig. 6b. CNT intimately connected with V6O13 nanosheets for faster electron transferring. Furthermore, CNTs and carbon fiber in the c-textile composed hierarchical conductive network for better electron conducting. The cyclability of V6O13 electrodes was shown in Fig. 4d. At the specific current of 300 mA g−1, the electrode with CNT coating could maintain 74% of the initial capacity of 170 mA h g−1 after 300 charge/discharge cycles, while the V6O13 electrode only retained 42% of its initial capacity. It outperformed most of the low dimensional mixed-valence vanadium oxides or their 3D nanostructure listed in Additional file 1: Table S2. The better cyclability of V6O13 electrode with CNT might be ascribed to the following reasons: (1) Reinforced with CNT, V6O13’s mechanical properties were improved. (2) Even if V6O13 nanostructure was broken during the discharge/charge process, it was still attached to CNT and could be electrochemically activated. (3) Self-segregation of V6O13 nanosheets was limited by the appearance of CNT. (4) CNT coating might be a valid barrier to alleviate side reaction of vanadium oxide with electrolyte, if any. Thus, CNT coating can be a facile alternative way to improve the conductivity of the 3D nanostructure, other than carbon coating and polymeric coating which usually require tremendous chemical synthesis work [14]. The overall electrochemical performance of V6O13 cathode was limited by the conductivity of carbon cloth, the Li diffusivity in V6O13 materials, and electron transfer between V6O13 nanostructures and the carbon cloth. In the future work, further improvement can be made in the following ways: (1) reducing the resistance of the carbon cloth substrate, (2) doping V6O13 with sulfur to improve its diffusivity of lithium ion, and (3) coating the V6O13 with conductive polymer coating.

Fig. 4
figure 4

Electrochemical performance of 3D free-standing V6O13 electrodes with/without CNT coating. a Cyclic voltammetry curves. b Galvanostatic charge/discharge curves. c Rate. d Cyclability performance of the two electrodes

Full size image
Fig. 5
figure 5

Microstructure of CNT-coated V6O13 electrode. a Low-resolution SEM image of the electrode. b, c High-resolution SEM image of the electrode showing CNT covering on the nanosheet and bridging between the nanosheets

Full size image
Fig. 6
figure 6

a Nyquist plots of V6O13 electrode with/without CNT coating. b Schematic of the transporting path for electrons in CNT-coated V6O13 electrode

Full size image

Conclusions

In summary, we successfully fabricated 3D free-standing V6O13 nanostructure composed of interconnected nanosheets via a facile seed-assisted hydrothermal process as a cathode for LIB. The electrode exhibited good electrochemical performance. It could be further improved by coating 3D V6O13 nanostructure with CNT, outperforming most of the mixed-valence vanadium oxides. Its excellent performance was due to its open 3D nanostructure and hierarchical conductive network composed of CNT in nanoscale and carbon fiber in microscale. The design of 3D nanostructure with the building block (e.g., nanowire, nanosheet) combined with constructing of the hierarchical conductive path by CNT coating can be extended to other electrode materials for better electrochemical performance.