A novel synthesis towards a vanadium pentoxide porous nanodisk film as a cathode material for advanced Li-ion hybrid capacitors

Abstract

A V₂O₅ porous nanodisk thin film is synthesized through a simple hydrothermal and subsequent VO₂ template oxidation strategy. For the first time, V₂O₅ is employed as a cathode rather than an anode to construct lithium-ion hybrid capacitors. This design effectively utilizes the intrinsic layered structure of V₂O₅ for facile Li⁺ intercalation and facilitates the charge balance with the capacitive electrode, enabling superior performance of the device.

---

With the increasing market demand of new energy vehicles and flexible electronics, the development of energy storage devices with high power and energy densities is particularly urgent.¹ Among various energy storage systems, lithium-ion batteries (LIBs) can provide high energy density due to faradaic electrode reactions, but have limited power density.^2–4 Compared with LIBs, supercapacitors show high power density and outstanding cycling stability due to the surface/near-surface charge storage behavior; however, they generally deliver low energy density.^4–6 To achieve high comprehensive performance, lithium-ion hybrid capacitors (LIHCs) combining the advantages of LIBs and supercapacitors were developed, which are assembled with one battery-type electrode and one capacitive electrode.^7–9 However, there remains a great challenge in realizing the kinetics balance and charge balance between the two electrodes of different energy storage mechanisms.^1,2 It is of utmost importance to develop a high-rate battery electrode with comparable capacity to match with the capacitive electrode in LIHCs.

Vanadium oxides possess rich varieties of structural motifs due to their different atomic configurations and multiple V valence states,^10–13 which provides superior Li ion storage capability.^14,15 In particular, vanadium pentoxide (V₂O₅) exhibits a layered structure and large interlayer spacing (4.37 Å) that benefit the accommodation of Li⁺.¹⁶ In previous studies, V₂O₅ was often employed as an anode to construct LIHCs. Unfortunately, when V₂O₅ is used as an anode, the crystal structure will be completely damaged during cycling due to the conversion reaction mechanism, resulting in poor cycling stability. In such a case, the benefits of the intrinsic layered structure cannot be utilized, which also leads to poor rate performance. In addition, like anodes of other transition metal oxides, the capacity (500–1000 mA h g⁻¹) of the V₂O₅ anode is much higher than that of the capacitive cathode,^17,18 which causes great difficulty in realizing the charge balance between the anode and cathode in LIHCs. Then, it is of great interest to design LIHCs with V₂O₅ as a cathode. In order to achieve high performance of the above proposed LIHC, it is still necessary to boost the rate performance of the V₂O₅ cathode. Therefore, designing a binder-free film electrode with a porous nanostructure may be one of the best choices, as such an electrode architecture can provide not only direct electron transport channels, but also shortened ion diffusion paths benefiting from the sufficient electrode/electrolyte contact interface.^19,20

Herein, we report a unique VO₂ template oxidation strategy to synthesize a V₂O₅ porous nanodisk film and further use it as a cathode to design a LIHC. The removal of residual organic species from the VO₂ structure via direct high temperature annealing in air gives rise to the formation of abundant pores in V₂O₅ nanodisks. The porous structure combined with the merits of a binder-free architecture enables good rate capability of the electrode. By pairing with a prelithiated activated carbon (LiAC) anode, the assembled thin-film LIHC device (the energy storage mechanism is elucidated in Fig. S1, ESI†) also delivers remarkable energy densities (146 W h kg⁻¹, 13 mW h cm⁻³) and high power densities. Furthermore, the device still works well when bended or even folded.

The precursor VO₂ nanosheet array was grown on a thin steel mesh current collector by a simple hydrothermal method, and the V₂O₅ nanodisk film was obtained via further oxidization of VO₂ in air at 550 °C (see the ESI† for details). Scanning electron microscopy (SEM) images show that the nanostructures are uniformly grown on the surface of each stem of the steel meshes (Fig. S2, ESI†). Compared to the precursor VO₂, the V₂O₅ film exhibits a different color with a clear structural transformation from thin nanosheets to thick nanodisks (Fig. 1a and b and insets). In addition, a highly porous structure of V₂O₅ can be detected by transmission electron microscopy (TEM), as shown in Fig. 1c. The lattice fringes with a spacing of ∼0.261 nm correspond well to the (310) crystal planes of V₂O₅. The N₂ adsorption–desorption isotherm in Fig. 1d displays the type-IV hysteresis loop, further indicating the existence of porous features. The pores are distributed with the sizes smaller than 5 nm, including both the mesopores and micropores (inset in Fig. 1d). To understand the formation mechanism of the pores, thermal gravimetric analysis (TGA) was performed (Fig. S3a, ESI†). In addition to the weight loss of ∼2.12 wt% below 250 °C that is attributed to the loss of surface-adsorbed and weakly bound water, a weight drop of ∼4.63% at around 270 and 350 °C with two noticeable exothermic peaks is observed, which is consistent with the TGA feature of oxalic acid.²¹ Therefore, organic species of oxalic acid was present in VO₂ sheets during the hydrothermal process, the removal of which is responsible for the porous structure within the V₂O₅ disks.

Fig. 1 Top-view SEM images of (a) precursor VO₂ and (b) V₂O₅ nanodisk film. The insets are optical pictures. (c) TEM image and (d) nitrogen adsorption/desorption isotherm and pore size distribution of V₂O₅. (e) XRD results of precursor VO₂ and the resulting V₂O₅. (f) XPS result of V₂O₅.

The crystal structure and component of the V₂O₅ film electrode was further characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy. As shown in Fig. 1e, the XRD peaks clearly indicate the complete transformation from triclinic VO₂ (JCPDS Card No. 19-1401) to orthorhombic V₂O₅ (JCPDS Card No. 41-1426) with a trace amount of V₄O₉ (not fully oxidized) after annealing the electrode in air. The skeleton bent vibration at 146 cm⁻¹ and the stretching vibration of vanadyl at 997 cm⁻¹ in the Raman spectrum are characteristic of the layered structure of V₂O₅ (Fig. S3b, ESI†).^22,23 The full XPS spectrum (Fig. S3c, ESI†) demonstrates the presence of V and O elements. After deconvolution, the asymmetric shape of the V2p band can be separated into two sets of valence states.¹⁰ In detail, the spin-orbital splitting components of V2p_3/2 and 2p_1/2 peaks are observed at 517.2 and 524.7 eV, respectively (Fig. 1f).^24,25 The V2p_3/2 peak of vanadium can be divided into two peaks at binding energies of 517.3 and 516.2 eV, respectively, assigned to V⁵⁺ and V⁴⁺.^26,27 The V2p_1/2 peak can also be divided into two binding energies of 524.9 and 523.6 eV attributed to V⁵⁺ and V⁴⁺, respectively.²⁸ Nevertheless, the oxide with V⁵⁺ is found to be the dominant phase.

Next, the electrochemical performance of the V₂O₅ nanodisk film electrode was measured. The charge–discharge profiles of the V₂O₅ electrode within 2.0 to 4.0 V (vs. Li/Li⁺) at 100 mA g⁻¹ are displayed in Fig. 2a. In the first discharge curve, there are obvious lithium insertion plateaus. Subsequently, the plateaus become invisible, indicating that the crystallinity decreases after the first lithiation/delithiation (Fig. S4, ESI†).²⁸ The differential capacity versus potential plots in Fig. 2b show three pairs of reversible small redox peaks after the first cycle. The three cathodic peaks at 3.39, 3.18 and 2.30 V represent the multistep insertion process of lithium ions into V₂O₅, corresponding to the phase transition of α-V₂O₅ to ε-Li_0.5V₂O₅, δ-LiV₂O₅, and γ-Li₂V₂O₅, respectively.²⁹ In reverse, the anodic peaks observed at 2.47, 3.32 and 3.43 V can be ascribed to the de-insertion of Li⁺ from V₂O₅.³⁰Fig. 2c and d further show the rate capability. The V₂O₅ electrode exhibits similar charge/discharge profiles at various current densities with high coulombic efficiencies. At 25 mA g⁻¹, the capacity is as high as ∼300 mA h g⁻¹. With the current density increasing 64 times from 25 to 1600 mA g⁻¹, a capacity of 104 mA h g⁻¹ can still be maintained, indicative of excellent rate performance. Our V₂O₅ electrode also shows good cycling stability, as shown in Fig. 2e. After 300 times cycling at 200 mA g⁻¹, the reversible capacity is still 162.9 mA h g⁻¹, which maintains 68% of the initial capacity. The capacity loss rate is approximately 0.11% per cycle, which is superior to those of other nanostructured V₂O₅ electrodes, such as V₂O₅ nanofibers (0.52% per cycle)³¹ and carbon supported V₂O₅ nanotubes (0.42% per cycle).²⁸

Fig. 2 (a) Typical charge/discharge profiles of the V₂O₅ electrode at 100 mA g⁻¹. (b) 1st and 2nd cycle differential capacity versus potential (V) plots. (c) Charge and discharge curves of the V₂O₅ electrode at different rates. (d) Rate performance. (e) Cycling performance at 200 mA g⁻¹.

To further explore the electrochemical merits of our V₂O₅ nanodisk film, an LiAC₍₋₎//V₂O₅₍₊₎ LIHC device using V₂O₅ as a cathode and LiAC as an anode was assembled and tested. The charge and discharge behavior of LiAC (KYP-50) was evaluated in a half cell prior to the construction of the LIHC. After prelithiation and formation of a stable solid electrolyte interface (Fig. S5a, ESI†),³² the LiAC electrode exhibits stable CVs and charge–discharge curves (Fig. 3a and b) in the potential range of 0.3–2.4 V (vs. Li/Li⁺). Fig. S5b (ESI†) shows the CVs of the LiAC at different sweep rates. The cathodic and anodic currents I_p are plotted with sweep rates (v) and shown in the inset of Fig. 3a. The results can be well fitted with a linear dependence of I_p with v at relatively low sweep rates, which indicates that the electrochemical process manifests a capacitive behavior⁶ (although this process is different from the case when AC is used as the cathode; Fig. S6, ESI†). Further charge/discharge testing at various current densities (Fig. 3c and d) reveals a good rate performance of the LiAC electrode. The reversible discharge capacity can reach 150 mA h g⁻¹ at 25 mA g⁻¹; at 1600 mA g⁻¹, a capacity of 70 mA h g⁻¹ can still be obtained. The LiAC electrode also displays good cycling stability at 200 mA g⁻¹. As shown in Fig. 3e, ∼80% of the initial capacity can be retained after 1000 continuous cycles.

Fig. 3 (a) CV curves of the LiAC anode at 0.1 mV s⁻¹. The inset shows the plots of the peak current density versus the sweep rate. (b) Typical charge/discharge profiles of the LiAC anode at 50 mA g⁻¹. (c) Charge and discharge curves of the LiAC anode. (d) Rate performance. (e) Cycling performance.

Fig. 4 illustrates the electrochemical performance of the assembled LIHC device within 0.01–3.7 V. Based on the half-cell results, the mass ratio of the anode to cathode is set as 2.3 : 1 to realize the charge balance. As shown in Fig. 4a–c and Fig. S7 (ESI†), our device always exhibits almost linear charge–discharge plots at different current densities, indicative of a capacitive behavior. Good rate performance is also achieved, with ∼40% of the initial capacity retained when the current is increased 32 times to 800 mA g⁻¹. The device can be subjected to continuous cycling with gradually increased current rates, and more importantly, when the current turns back to the initial value of 25 mA g⁻¹, the capacity is almost recovered. Fig. 4d demonstrates the good cycling stability of the device. The charge/discharge capacity shows small loss after 100 cycles, with high coulombic efficiencies of 95–100%.

Fig. 4 (a) Typical charge/discharge profiles of our LiAC₍₋₎//V₂O₅₍₊₎ LIHC. (b) Rate performance. (c) Charge/discharge profiles at different rates. (d) Cycling performance at 50 mA g⁻¹. The inset shows the device with different shape deformations powering LED indicators. (e and f) Ragone plots.

Our V₂O₅ film was grown tightly on a thin steel mesh without any binders or additives, which was not even able to peel off during repeated bending. This feature encourages us to assemble a soft-packing LIHC device to check the flexibility. The inset in Fig. 4d shows the optical images of our prototype device, which can efficiently power a 3.0 V white LED indicator under severe bending and folding conditions. The potential of our LIHC for energy storage application can be further manifested by its energy and power densities. Fig. 4e shows the Ragone plot of gravimetric energy density versus power density of our device, including data from previously reported systems for comparison. In general, the gravimetric energy densities of our LIHC (146 W h kg⁻¹ at 24 W kg⁻¹; 78 W h kg⁻¹ at 252 W kg⁻¹) are much larger than those of LIHCs using vanadium oxides as the anode, such as V₂O₅–CNT//AC (40 W h kg⁻¹ at 200 W kg⁻¹)³³ and V₂O₅//CNT (18 W h kg⁻¹ at 42 W kg⁻¹).³⁴ The energy densities are also superior to many other LIHCs such as Nb₂O₅–CNT//AC (40 W h kg⁻¹ at 20 W kg⁻¹),³⁵ Nb₂O₅ core-cell//AC (63 W h kg⁻¹ at 70 W kg⁻¹),³⁶ LiTi₂(PO₄)₃//AC (24 W h kg⁻¹ at 200 W kg⁻¹)³⁷ and Li₄Ti₅O₁₂//AC (33 W h kg⁻¹ at 50 W kg⁻¹).³⁸ At a high power density of 1396 W kg⁻¹, the energy density of our device is still as high as 35 W h kg⁻¹, better than conventional supercapacitors. Furthermore, our device delivers high volumetric energy densities of 13 mW h cm⁻³ at 3 mW cm⁻³ and 4.2 mW h cm⁻³ at 71.9 mW cm⁻³ (Fig. 4f), which are higher than the values of previous thin-film or flexible LIHCs, such as Li₄Ti₅O₁₂//CNT (∼2.5 mW h cm⁻³ at 45 mW cm⁻³)³⁹ and WO_3−x@MoO_3−x//PANI (∼1.9 mW h cm⁻³ at 40 mW cm⁻³).⁴⁰ The values are even better than those of commercial Li thin-film batteries and supercapacitors.

In summary, a V₂O₅ porous nanodisk film is synthesized directly on a current collector by a simple hydrothermal and subsequent template oxidation method. The V₂O₅ electrode exhibits high capacity and good rate performance, which is attributed to the direct electron transport along the ordered electrode architecture and rapid Li-ion migration within the porous nanodisk structure. The resulting LIHC device delivers high energy and power densities as well as good flexibility. This work highlights the importance of using the intrinsic layered structure of materials for developing hybrid capacitors.

This work was supported by the National Natural Science Foundation of China (Grant No. 51872104, 51672205 and 51972257), the National Key R&D Program of China (Grant No. 2016YFA0202602), and the Natural Science Foundation of Hubei Province (2018CFB581).

Conflicts of interest

There are no conflicts to declare.

References

1. W. H. Zuo, R. Z. Li, C. Zhou, Y. Y. Li, J. L. Xia and J. P. Liu, Adv. Sci., 2017, 4, 1600539.
2. H. Kim, M.-Y. Cho, M.-H. Kim, K.-Y. Park, H. Gwon, Y. Lee, K. C. Roh and K. Kang, Adv. Energy Mater., 2013, 3, 1500–1506.
3. X. Zhao, H. E. Wang, J. Cao, W. Cai and J. Sui, Chem. Commun., 2017, 53, 10723–10726.
4. Q. Wang, Z. H. Wen and J. H. Li, Adv. Funct. Mater., 2006, 16, 2141–2146.
5. S. W. Lee, N. Yabuuchi, B. M. Gallant, S. Chen, B. S. Kim, P. T. Hammond and Y. Shao-Horn, Nat. Nanotechnol., 2010, 5, 531–537.
6. B. H. Deng, T. Y. Lei, W. H. Zhu, L. Xiao and J. P. Liu, Adv. Funct. Mater., 2018, 28, 1704330.
7. Y. Q. Jiang and J. P. Liu, Energy Environ. Mater., 2019, 2, 30–37.
8. G. Qu, B. B. Tian, C. L. Su, Y. W. Tang and Y. Li, Chem. Commun., 2018, 54, 10355–10358.
9. Y. Xu, E. M. Lotfabad, H. Wang, B. Farbod, Z. Xu, A. Kohandehghan and D. Mitlin, Chem. Commun., 2013, 49, 8973–8975.
10. H. E. Wang, X. Zhao, K. Yin, Y. Li, L. Chen, X. Yang, W. Zhang, B. L. Su and G. Cao, ACS Appl. Mater. Interfaces, 2017, 9, 43665–43673.
11. W. H. Zhu, R. Z. Li, P. Xu, Y. Y. Li and J. P. Liu, J. Mater. Chem. A, 2017, 5, 22216–22223.
12. Y. Zhang, X. Jing, Q. Wang, J. Zheng, H. Jiang and C. Meng, Dalton Trans., 2017, 46, 15048–15058.
13. X. Liu, J. Zeng, H. Yang, K. Zhou and D. Pan, RSC Adv., 2018, 8, 4014–4031.
14. D. Graf, J. Schläfer, S. Garbe, A. Klein and S. Mathur, Chem. Mater., 2017, 29, 5877–5885.
15. C. Wu, F. Feng and Y. Xie, Chem. Soc. Rev., 2013, 42, 5157–5183.
16. M. Liu, B. Su, Y. Tang, X. Jiang and A. Yu, Adv. Energy Mater., 2017, 7, 1700885.
17. F. Cheng and J. Chen, J. Mater. Chem., 2011, 21, 9841–9848.
18. Y. Zhao, X. Li, B. Yan, D. Xiong, D. Li, S. Lawes and X. Sun, Adv. Energy Mater., 2016, 6, 1502175.
19. H. Gwon, J. Hong, H. Kim, D.-H. Seo, S. Jeon and K. Kang, Energy Environ. Sci., 2014, 7, 538–551.
20. J. Jiang, Y. Li, J. P. Liu, X. Huang, C. Yuan and X. W. Lou, Adv. Mater., 2012, 24, 5166–5180.
21. M. R. Udupa, Thermochim. Acta, 1981, 51, 169–173.
22. Q. Su, X. Q. Liu, H. L. Ma, Y. P. Guo and Y. Y. Wang, J. Solid State Electrochem., 2008, 12, 919–923.
23. P. Chen, G. Zheng, G. Guo, Z. Wang, J. Tang, S. Li, Z. Wen, S. Ji and J. Sun, J. Alloys Compd., 2019, 784, 574–583.
24. Z. Li, G. Liu, M. Guo, L.-X. Ding, S. Wang and H. Wang, Electrochim. Acta, 2015, 173, 131–138.
25. X. Liang, G. Gao, Y. Liu, T. Zhang and G. Wu, J. Alloys Compd., 2017, 715, 374–383.
26. Q. Wei, Z. Jiang, S. Tan, Q. Li, L. Huang, M. Yan, L. Zhou, Q. An and L. Mai, ACS Appl. Mater. Interfaces, 2015, 7, 18211–18217.
27. F. Xiao, X. Song, Z. Li, H. Zhang, L. Zhang, G. Lei, Q. Xiao, Z. Hu and Y. Ding, J. Mater. Chem. A, 2017, 5, 17432–17441.
28. X. Liang, G. Gao, S. Feng, Y. Du and G. Wu, J. Alloys Compd., 2019, 772, 429–437.
29. X. F. Zhang, K. X. Wang, X. Wei and J. S. Chen, Chem. Mater., 2011, 23, 5290–5292.
30. Y. Li, J. Yao, E. Uchaker, J. Yang, Y. Huang, M. Zhang and G. Cao, Adv. Energy Mater., 2013, 3, 1171–1175.
31. Y. L. Cheah, N. Gupta, S. S. Pramana, V. Aravindan, G. Wee and M. Srinivasan, J. Power Sources, 2011, 196, 6465–6472.
32. X.-F. Luo, C.-H. Yang, Y.-Y. Peng, N.-W. Pu, M.-D. Ger, C.-T. Hsieh and J.-K. Chang, J. Mater. Chem. A, 2015, 3, 10320–10326.
33. Z. Chen, V. Augustyn, J. Wen, Y. Zhang, M. Shen, B. Dunn and Y. Lu, Adv. Mater., 2011, 23, 791–795.
34. V. Aravindan, Y. L. Cheah, W. F. Mak, G. Wee, B. V. R. Chowdari and S. Madhavi, ChemPlusChem, 2012, 77, 570–575.
35. X. Wang, G. Li, Z. Chen, V. Augustyn, X. Ma, G. Wang, B. Dunn and Y. Lu, Adv. Energy Mater., 2011, 1, 1089–1093.
36. E. Lim, C. Jo, H. Kim, M. Kim, Y. Mun, J. Chun, Y. Ye, J. Hwang, K. Ha and K. J. A. N. Roh, ACS Nano, 2015, 9, 7497–7505.
37. J.-Y. Luo and Y.-Y. Xia, J. Power Sources, 2009, 186, 224–227.
38. H.-G. Jung, N. Venugopal, B. Scrosati and Y.-K. Sun, J. Power Sources, 2013, 221, 266–271.
39. W. H. Zuo, C. Wang and J. P. Liu, Sci. Rep., 2015, 5, 7780.
40. X. Xiao, T. Ding, L. Yuan, Y. Shen, Q. Zhong, Xi. Zhang, Y. Cao, B. Hu, T. Zhai, L. Gong, J. Chen, Y. Tong, J. Zhou and Z. L. Wang, Adv. Energy Mater., 2012, 2, 1328.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details. See DOI: 10.1039/c9cc06958e

---