Preparation and electrochemical performance of Mo₆V₉O₄₀ nanorods as cathode materials for Li batteries

Abstract

Mo₆V₉O₄₀ nanorods were prepared through a sol–gel route followed by short-time calcination, and tested as cathode materials for lithium batteries. The Mo₆V₉O₄₀ nanorods demonstrated good cyclic stability and rate performance at deep discharge (discharged to 1.5 V), which were attributed to the enhanced structural flexibility and the 1D structure.

---

Introduction

Searching for cathode materials of Li-based batteries with high energy densities has been a key mission in recent years.^1,2 Among the various candidates, vanadium based cathode materials (such as LiV₃O₈ and V₂O₅) and molybdenum oxides have been intensively studied due to their high capacity, easy preparation, and low cost.^3–9 Li ions could be reversibly intercalated/deintercalated into the inter layers of oxides such as V₂O₅ and MoO₃. V₂O₅ presents a layered structure, which easily undergoes phase transitions from α-Li_xV₂O₅ (x < 0.01), ε-Li_xV₂O₅ (0.35 < x< 0.7), and ε-Li_xV₂O₅ to δ-Li_xV₂O₅ (x < 1), and then δ-Li_xV₂O₅ to γ-Li_xV₂O₅ (x > 1) when the discharge cutoff potential is set to 2 V (vs. Li/Li⁺).¹⁰ These phase transitions are reversible, and the layered structure is not destroyed. However, when discharged to 1.5 V, the transition from γ-Li_xV₂O₅ to ω-Li_xV₂O₅ (x > 2) occurs, which is irreversible and leads to poor electrochemical performance.¹¹ The hybrid oxides, molybdenum–vanadium based oxides, have been reported to be more sustainable to over discharge (to 1.5 V), which include solid solution (V_1−yMo_y)₂O₅ (0 ≤ y ≤ 0.3) and Mo₆V₉O₄₀.^12–14 The molybdenum and vanadium ions are distributed disorderly in (V_1−yMo_y)₂O₅ and regularly in Mo₆V₉O₄₀. Mo₆V₉O₄₀ was reported as MoV₂O₈ initially, and the prepared MoV₂O₈ was invariably accompanied with V₂O₅ impurities according to the previous reports.^15,16 Introduction of Mo to the structure increases the length of M–O bonds, which is more flexible for Li ion insertion/extraction.¹⁷ In addition, substituting Mo⁶⁺ for V⁵⁺ results in the presence of V⁴⁺ and increases the electronic conductivity.

Molybdenum and vanadium mixed oxides are usually prepared through solid state reactions above 600 °C with extremely long calcination time (over 50 h) in the previous reports.^17,18 However, the long heating time at high temperatures degrades the morphology and results in large particles with several microns in diameter, which does not benefit the electrochemical performance and large scale production.¹⁸ In this work, Mo₆V₉O₄₀ nanorods were prepared by a facile sol–gel method. The gel was firstly heated in Ar atmosphere by a two-step calcine process to get the precursor, in which vanadium and molybdenum elements were dispersed uniformly in excessive amorphous carbon. The Mo₆V₉O₄₀ nanorods were prepared by caclining the obtained precursor in air, and the time for phase formation was reduced to 2 h. The preparation time was remarkably decreased. The Mo₆V₉O₄₀ nanorods exhibited good cyclic stability and rate performance.

Experimental

Material preparation

The Mo₆V₉O₄₀ samples were prepared by a sol–gel method described elsewhere with minor modifications.^19,20 In brief, 0.0883 g ammonium molybdate tetrahydrate, 0.117 g ammonium metavanadate, 1 g citric acid, and 5 g urea were dissolved in 200 ml water–ethanol mixed solvent with agitation under 75 °C to form a homogenous solution. Here, the amount of ammonium metavanadate was excessive to promote the generation of Mo₆V₉O₄₀. The color of the solution changed from orange to blue due to the reduction and complexation of V ions. After the excess solvent was evaporated, the formed gel was dried under 100 °C in air. The gel was preheated at 350 °C for 4 h and then calcined at 650 °C for 10 h in flowing argon. The obtained precursor was finally calcined in air under different temperatures to get Mo₆V₉O₄₀ samples.

Material characterization

Powder X-ray diffraction (XRD) patterns were recorded on a D/MAX III diffractometer with Cu Kα radiation. Morphologies of the samples were observed on field emission scanning electron microscope (FESEM, JEOL-JSM7500) and transmission electron microscope (TEM, Tecnai G2 Spirit). The composition of the samples was confirmed by the energy dispersive X-ray spectroscopy (EDX) attached to the FESEM instrument.

Electrochemical tests

For the electrode preparation, Mo₆V₉O₄₀ (80 wt%), acetylene black (10 wt%), and polyvinylidene fluoride (PVDF) binder (10 wt%) in N-methylpyrrolidone were mixed, and the obtained homogenous slurry was pasted on Al foil, followed by drying in vacuum at 110 °C. Coin cells (CR2025) were fabricated in an argon-filled glove box with lithium metal as the counter electrode, and LiPF₆ (1 M) in ethylene carbonate/ethyl methyl carbonate/dimethyl carbonate (EC/EMC/DMC, 1 : 1 : 1 vol.%) as the electrolyte. Galvanostatic charge/discharge tests were performed on a Land CT2001 battery tester in the voltage range of 1.5–4 V at room temperature. Cyclic voltammogram (CV) tests were performed under a Zahner-Elektrik IM6e electrochemical workstation in the voltage range of 1.5–4 V at a scan rate of 0.1 mV s⁻¹.

Results and discussion

To decrease the agglomeration of the Mo₆V₉O₄₀ materials, the preparation temperature was optimized and the calcining time was limited to 2 h. The XRD patterns of the samples obtained at different temperatures are shown in Fig. 1. When calcined at 500 °C, the samples are composed of vanadium oxides and molybdenum oxides at low valence. The complete monoclinic Mo₆V₉O₄₀ phase formed at 550 °C with the space group of C121 (JCPDS#74-1510), and the impurity could be indexed as V₂O₅. The Mo₆V₉O₄₀ diffraction peaks are unchanged when calcined at higher temperatures. The SEM images of the samples obtained at different temperatures are shown in Fig. 2. The particles of the sample calcined at 500 °C are bulk and irregular. When calcined at 550 °C, the sample is composed of uniform nanorods with the diameters of ∼200 nm. The particle size increases obviously when the calcined temperature increased to 600 °C, and the cuboid structure is observed. The cuboid structure is more common when calcined at 650 °C, and the diameters increase to several microns. The particle size and morphology are very sensitive to the calcination temperature, and large particles block the Li ion diffusion. Therefore, the Mo₆V₉O₄₀ sample calcinated at 550 °C was chosen for the following studies.

Fig. 1 XRD patterns of the Mo₆V₉O₄₀ samples prepared under different calcination temperatures (■ V₂O₅ impurity).

Fig. 2 SEM images of the Mo₆V₉O₄₀ samples under different calcination temperatures: (a) 500 °C, (b) 550 °C, (c) 600 °C, and (d) 650 °C.

The microstructures of the Mo₆V₉O₄₀ sample calcinated at 550 °C are shown in Fig. 3. The Mo₆V₉O₄₀ nanorods disperse consistently and the length is lower than 1 μm. The elemental mappings of the Mo₆V₉O₄₀ nanorods are shown in Fig. S1 (ESI†), and the Mo, V, and O elements are distributed homogeneously. According to the EDX result, the atomic ratio of Mo and V is close to 1 : 1.5. The HRTEM images in Fig. 3d and e further confirm the one-dimensional (1D) rod shape of the sample, which grows regularly. Fig. 3f shows the lattice of the nanorods in Fig. 3e. The interlayer spacing is 0.41 nm, determined as the (111) crystal planes.

Fig. 3 (a and b) SEM images, (c) EDX spectrum, and (d–f) TEM images of the Mo₆V₉O₄₀ nanorods calcined at 550 °C.

The electrochemical performances were evaluated in coin cells with Mo₆V₉O₄₀ as the cathode and metallic lithium as the anode. The CV curves are shown in Fig. 4a. The position of the cathodic peaks in the first cycle is similar to that of V₂O₅ tested between 4.0 and 1.5 V, indicating the multiple phase transition.²¹ The peaks at 3.1, 2.9, 2.2, and 1.8 V correspond to the α–ε–δ–γ–ω phase transition of V₂O₅, although an additional peak at 2.65 V is observed, which may be due to the introduction of Mo. The small cathodic peak at ∼2.45 V is attributed to the V₂O₅ impurity due to the small phase transition, which was also found in previous reports.^22,23 The cathodic peaks in the first cycle disappear in the following cycles, and two new pairs of anodic peaks and cathodic peaks emerge, which are stable in the subsequent cycles. Fig. 4b shows the galvanostatic charge/discharge curves at different cycles at the current density of 100 mA g⁻¹. The initial discharge curve is also special, and the discharge plateaus are consistent with the CV curves. The discharge capacity is 369 mA h g⁻¹, corresponding to the intercalation of 1.5 Li⁺ ions per metal ion. The ratio of the discharge capacity corresponding to the formation of ω phase (the 2 V plateau) is obviously larger than that of V₂O₅. The ω phase of Mo₆V₉O₄₀ is formed not only by the intercalation of Li⁺ ions in the γ phase, but also in the δ phase and even ε phase.²¹ The ω phase of Mo₆V₉O₄₀ is more reversible than the ω phase of V₂O₅, in which Li⁺ ions could be deintercalated partly.¹⁷ Then part of Li⁺ ions are deintercalated in the charge process because the phase transition blocks the diffusion of part Li⁺ ions. The charge/discharge curves keep stable and the Li⁺ ions are reversibly inserted/extracted in the following cycles.

Fig. 4 (a) CV curves, (b) galvanostatic charge/discharge curves at different cycles at 100 mA g⁻¹, (c) cyclic performance at 100 mA g⁻¹ and 1 A g⁻¹, and (d) cyclic performance at different rates of the Mo₆V₉O₄₀ nanorods.

The cyclic performance of the Mo₆V₉O₄₀ nanorods is shown in Fig. 4c. The discharge capacity at the 2nd cycle is 276.2 mA h g⁻¹ at 100 mA g⁻¹ and retains 214.8 mA h g⁻¹ after 50 cycles. When tested at 1 A g⁻¹, the discharge capacity at the 2nd cycle is 212 mA h g⁻¹ and retains 181.8 mA h g⁻¹ after 50 cycles. The Mo₆V₉O₄₀ nanorods were also tested at other current densities (Fig. 4d). The discharge capacities are 239.4, 190.8, and 181.1 mA h g⁻¹ at 0.3, 0.5, and 0.8 A g⁻¹, respectively. When the current density returns to 100 mA g⁻¹, the discharge capacity recovers to 230.4 mA h g⁻¹. The Mo₆V₉O₄₀ nanorods exhibited good cyclic stability and rate performance compared with the previous reports.^17,24 Uchiyama et al. synthesized Mo₆V₉O₄₀ by solid state reactions followed by quenching to room temperature, and the capacity retention was about 50% after 32 cycles.¹⁷ Delmas and Cognac-Auradou synthesized Mo₆V₉O₄₀ by a long time calcination (70 h), and the capacity retention was about 68% after 50 cycles.²⁴ Fig. S2 (ESI†) shows the Coulombic efficiency when tested at 100 mA g⁻¹. The Coulombic efficiency approaches 100% after the first cycle, indicating that no side reactions occur. The excellent performances at deep discharge are due to the enhanced structural flexibility resulting from the change in M–O bond lengths caused by substituting Mo for V. The 1D structure also facilitates the Li⁺ ion diffusion and enhances the rate performance.

Conclusions

In summary, we have developed a facile sol–gel method to synthesize Mo₆V₉O₄₀ nanorods. The Mo₆V₉O₄₀ nanorods are 200 nm in diameter, and their lengths are 500 nm to 1 μm. The electrochemical performances of the Mo₆V₉O₄₀ nanorods as cathode materials for Li batteries were investigated and showed high electrochemical stability at deep discharge, with a reversible capacity up to 276.2 mA h g⁻¹ at 100 mA g⁻¹ and a high C-rate performance with a discharge capacity 212 mA h g⁻¹ and retain 181.8 mA h g⁻¹ after 50 cycles at 1 A g⁻¹. The good cyclic stability and rate performance are attributed to the enhanced structural flexibility caused by substitution of Mo for V and the 1D structure, which allows for fast Li⁺ ion diffusion.

Acknowledgements

This work was supported by Tianjin Municipal Science and Technology Commission (14ZCZDGX00039), NSFC (21421001) and MOE Innovation Team (IRT13022) in China.

Notes and references

1. B. Dunn, H. Kamath and J. M. Tarascon, Science, 2011, 334, 928.
2. M. Hu, X. Pang and Z. Zhou, J. Power Sources, 2013, 237, 229.
3. G. Yang, G. Wang and W. Hou, J. Phys. Chem. B, 2005, 109, 11186.
4. X. Chen, H. Zhu, Y.-C. Chen, Y. Shang, A. Cao, L. Hu and G. W. Rubloff, ACS Nano, 2012, 6, 7948.
5. A.-M. Cao, J.-S. Hu, H.-P. Liang and L.-J. Wan, Angew. Chem., Int. Ed., 2005, 44, 4391.
6. C. H. Han, M. Y. Yan, L. Q. Mai, X. C. Tian, L. Xu, X. Xu, Q. Y. An, Y. L. Zhao, X. Y. Ma and J. L. Xie, Nano Energy, 2013, 2, 916.
7. H. B. Wu, A. Pan, H. H. Hng and X. W. Lou, Adv. Funct. Mater., 2013, 23, 5669.
8. L. Q. Mai, B. Hu, W. Chen, Y. Y. Qi, C. S. Lao, R. S. Yang, Y. Dai and Z. L. Wang, Adv. Mater., 2007, 19, 3712.
9. K. Sakaushi, J. Thomas, S. Kaskel and J. Eckert, Chem. Mater., 2013, 25, 2557.
10. E. Uchaker, N. Zhou, Y. Li and G. Cao, J. Phys. Chem. C, 2013, 117, 1621.
11. A. Sakunthala, M. V. Reddy, S. Selvasekarapandian, B. V. R. Chowdari and P. C. Selvin, Energy Environ. Sci., 2011, 4, 1712.
12. P. Fiordiponti, M. Pasquali, G. Pistoia and F. Rodante, J. Power Sources, 1981, 7, 133.
13. M. Pasquali, G. Pistoia and F. Rodante, J. Power Sources, 1981, 7, 145.
14. A. Tranchant and R. Messina, J. Power Sources, 1988, 24, 85.
15. Y. Muranushi, T. Miura, T. Kishi and T. Nagai, J. Power Sources, 1987, 20, 187.
16. W. Kaveevivitchai and A. J. Jacobson, Chem. Mater., 2013, 25, 2708.
17. M. Uchiyama, S. Slane, E. Plichta and M. Salomon, J. Electrochem. Soc., 1989, 136, 36.
18. P. Fiordiponti, M. Pasquali, G. Pistoia and F. Rodante, J. Power Sources, 1981, 7, 133.
19. L. Su, Z. Zhou and P. Shen, J. Phys. Chem. C, 2012, 116, 23974.
20. M. Hu, Y. Tian, L. Su, J. Wei and Z. Zhou, ACS Appl. Mater. Interfaces, 2013, 5, 12185.
21. X. Zhou, G. Wu, J. Wu, H. Yang, J. Wang, G. Gao, R. Cai and Q. Yan, J. Mater. Chem. A, 2013, 1, 15459.
22. Y. N. Ko, Y. Chan Kang and S. B. Park, Nanoscale, 2013, 5, 8899.
23. J. Shao, X. Li, Z. Wan, L. Zhang, Y. Ding, L. Zhang, Q. Qu and H. Zheng, ACS Appl. Mater. Interfaces, 2013, 5, 7671.
24. C. Delmas and H. Cognac-Auradou, J. Power Sources, 1995, 54, 406.

---

Footnote

† Electronic supplementary information (ESI) available: Elemental mappings of Mo₆V₉O₄₀ nanorods, and Coulombic efficiency cycled at 100 mA g⁻¹. See DOI: 10.1039/c4ra16335d

---