Formation of tungsten trioxide with hierarchical architectures arranged by tiny nanorods for lithium ion batteries

Abstract

WO₃ with hierarchical flower-like architectures has been obtained by calcination of WO₃·0.33H₂O, which is initially prepared via a hydrothermal method with formic acid as a structure directing agent. For the hierarchical flower-like structure, each flower petal is composed of a number of tiny nanorods and the smooth degree of the surface is tuned by the additive amount of HCOOH. As the hierarchical architecture shortens the diffusion paths of the electrolyte ions, improves the dynamic performance and supplies more active sites, the samples exhibit high discharge capacities and long cycling life. Especially, high discharge capacities of 1284.5 mA h g⁻¹, 904.1 mA h g⁻¹, 829.3 mA h g⁻¹, 576.8 mA h g⁻¹ and 861 mA h g⁻¹ are achieved when the current densities are 100 mA g⁻¹, 200 mA g⁻¹, 250 mA g⁻¹, 500 mA g⁻¹ and 100 mA g⁻¹, respectively.

---

1. Introduction

As one member of the metal oxides, which are the key ingredients for the development of many advanced functional materials and smart devices, tungsten trioxide (WO₃) with promising physical and chemical properties draws much attention and has been widely investigated.^1–8 With the change of ambient temperature, it always materializes with different crystal structures, such as monoclinic, triclinic, orthorhombic, tetragonal, cubic and hexagonal.⁹ In addition, WO₃ hydrates (WO₃·nH₂O (n = 0.33, 0.5, 1, 2)) or tungstic acid usually act as precursors to obtain the desired crystal phase of WO₃ by calcination.^10–12 Therefore, there is an effective approach to achieve WO₃ with a hierarchical architecture by precisely controlling the morphologies of WO₃·nH₂O.

To the best of our knowledge, the synthesis of WO₃·nH₂O is generally performed in a liquid-phase synthesis system, and a series of hierarchical WO₃·nH₂O, including flower-like, hollow, octahedral, and nanodisc structures and so on,^13–17 were recently prepared via a microwave-assisted solvothermal route,¹⁸ hydrothermal method,^19,20 solution-based colloidal approach²¹ and electrochemical anodization.²² The hydrothermal method is a conventional and accepted way to prepare materials with the advantages of being facile, cost-effective and easily controlled.

In this article, hexagonal WO₃ (h-WO₃) with hierarchical flower-like structures has been prepared with a combination of a hydrothermal method and a thermal treatment. Each flower petal is composed of a number of tiny nanorods and the smooth degree of the surface is tuned by the additive amount of HCOOH. In addition, h-WO₃ has trigonal cavities and hexagonal windows formed by WO₆ octahedra.²³ This unique tunnel structure benefits the insertion/extraction of cations with small size, resulting in a host for interactions for obtaining hexagonal tungsten bronzes M_xWO₃ (M = Li⁺, Na⁺, K⁺, etc.).²⁴ Therefore, excellent electrochemical performances are achieved as investigated for use as an electrode material for lithium ion batteries. Meanwhile, the hierarchical flower-like architecture also largely contributes to superior discharge capacities and a long cycling life.

2. Experimental

2.1 Synthesis

All the chemicals were of analytical grade and used without further purification. For a typical synthetic process, 1 mmol Na₂WO₄·2H₂O was dissolved in 18 mL deionized water under stirring, then 6 mL formic acid (HCOOH) was added into the above mixture. After stirring for another 10 min, the blended solution was transferred into 33 mL Teflon lined stainless steel autoclaves for hydrothermal reaction at 180 °C for 12 h. Finally, the precipitate was washed with deionized water and ethanol several times and dried at 60 °C for 10 h under vacuum. Experiments with 4 mL and 12 mL formic acid were also carried out, meanwhile the volume of deionized water was changed to 20 mL and 12 mL. The three as-prepared products with increasing formic acid (4 mL, 6 mL and 12 mL) were further processed by calcination at 350 °C for 3 h, and labelled S-1, S-2 and S-3, respectively.

2.2 Characterization

The crystalline structures of the as-prepared samples were characterized by X-ray diffraction (XRD) spectra (Rigaku D/Max-2500, Cu Kα radiation, λ = 0.1518 nm). The morphologies were detected by scanning electron microscopy (SEM) on a JEOL JSM-6700F (field emission) scanning electron microscope, and transmission electron microscopy (TEM) on a Tecnai G2 F20 TEM.

2.3 Electrochemical measurements

The working electrodes were fabricated by mixing active material (hierarchical flower-like h-WO₃), acetylene black and binder (PVDF) in a weight ratio of 80 : 10 : 10, and then coating on Cu foil. The mass of the active materials in the prepared electrodes was 1–2 mg. The lithium foil served as the counter and reference electrodes. For the electrolyte, 1 mol dm⁻³ solution of LiPF₆ dissolved in ethylene carbonate (EC), ethylene methyl carbonate (EMC) and dimethyl carbonate (DMC) (1 : 1 : 1 in volume) was used and the volume of electrolyte was ∼0.2 mL in each cell. The assembly of the testing cells was carried out in an argon filled glovebox, where water and oxygen concentrations were kept at less than 5 ppm. Galvanostatic charge/discharge measurements were operated on a Land CT2001 automatic battery tester in a voltage range of 0.01–3 V. Cyclic Voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed on a CHI660B electrochemical workstation. All the tests were performed at room temperature.

3. Results and discussions

Taking the typical experiment as an example, the experiment was conducted with 6 mL HCOOH at 180 °C for 12 h. The phase and purity of the as-prepared product was examined by powder X-ray diffraction measurements, as shown in Fig. S1a.† Obviously, all the diffraction peaks can be indexed to WO₃·0.33H₂O with hexagonal structure (h-WO₃·0.33H₂O) with lattice constants a = 7.285 Å, and c = 3.883 Å (JCPDS card 35-1001), apart from the peaks at 18° assigned to a rhombohedral structure of WO₃·0.33H₂O (o-WO₃·0.33H₂O, JCPDS card 87-1203). The strong and sharp diffraction peaks indicate the good crystallinity of the as-synthesized product. To further validate the existence of crystal water, thermogravimetric analysis (TG) was performed in an air atmosphere ranging from room temperature to 500 °C. As shown in Fig. S1b,† there is a small mass loss below 100 °C (0.3%), which can be ascribed to the volatilization of absorbed water. With the increase of temperature, a sharp mass loss of 2.6% is observed ranging from 100 °C to 300 °C, which is corresponds well to the theoretical value of 2.5% for WO₃·0.33H₂O.

The measurements using scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed to characterize the morphology and structures of the as-prepared product. As shown in Fig. 1a, the low magnification SEM image exhibits a uniformly dispersed hierarchical flower-like microstructure with an average size of ∼2 μm. Meanwhile, the high magnification SEM image shown in Fig. 1b shows with the details of the flower-like architecture. Interestingly, each petal with a rough surface is composed of tiny units, which will be further validated by TEM measurements. The low magnification TEM image in Fig. 1c exhibits a panoramic picture of the hierarchical flower-like structure. Meanwhile, the enlarged views of the flower petal marked by the rectangle and shown in Fig. 1c1 and c2 obviously show that the petal is formed of a number of tiny nanorods, which confirms the SEM observation well. Fig. 1d shows an HRTEM image and the fast Fourier transformation (FFT) pattern is shown as inset. The measured distance is 0.616 nm between every two adjacent lattice fringes, which corresponds to the (100) lattice plane and indicates the growth direction along [100].

Fig. 1 (a and b) The low and high magnification SEM, (c) TEM and (d) HRTEM images of the sample prepared with 6 mL HCOOH at 180 °C for 12 h (c1 and c2 are enlarged views of portions of (c) as signed by rectangles, and the inset in (d) is the FFT pattern).

With regard to the evolution of the hierarchical flower-like architecture of WO₃·0.33H₂O, time-dependent experiments were carried out, during which samples were collected at different time intervals (0.5 h, 1 h, 2 h, 5 h and 8 h). The SEM images and the corresponding XRD pattern are shown in Fig. 2. The diffraction peaks of the products collected at different time intervals in Fig. 2f are perfectly consistent with the final product shown in Fig. S1a.† At the beginning of the reaction (0.5 h), there are only scattered nanoparticles and some sheets observed in Fig. 2a. By prolonging the reaction time to 1 h, a number of nanoplates aggregated with nanoparticles are observed, as shown in Fig. 2b. As the reaction time further proceeds to 2 h (Fig. 2c), and then 4 h (Fig. 2d), the hierarchical flower-like architecture emerges, and is assembled by the sheets. As the reaction continues for 8 h, a structure similar to the final hierarchical flower-like structure forms as shown in Fig. 2e.

Fig. 2 SEM images of the samples synthesized at 180 °C for (a) 0.5 h, (b) 1 h, (c) 2 h, (d) 5 h and (e) 8 h, and (f) the corresponding XRD pattern.

Based on the observation of the above SEM images, shown in Fig. 2, a formation mechanism of the hierarchical flower-like WO₃·0.33H₂O is proposed and shown in Fig. 3. As illustrated obviously, time is the most important controlling factor in the “oriented attachment” process.^25–27 HCOOH is a weak acid (pK_a = 3.75) and its ionization equilibrium is shown as eqn (1). As this was introduced into the reaction system with WO₄²⁻, eqn (2) took place and numbers of H₂WO₄ nuclei formed and aggregated quickly by an oriented attachment mechanism to minimize the total surface free energy of the system, which perfectly follows Gibbs law.²⁸ Then dehydration of H₂WO₄ particles occurred under the conditions of high temperature and high pressure with the formation of WO₃·0.33H₂O nanoparticles, as shown in eqn (3). With time extension, the later generated WO₃·0.33H₂O tiny nanoparticles further nucleated on the sheets and grew, resulting in the prototype hierarchical flower-like architectures. Eventually, an Ostwald ripening process occurred with the formation of defined hierarchical flower-like structures.

HCOOH ⇌ H⁺ + HCOO⁻ (1)

WO₄²⁻ + 2H⁺ → H₂WO₄ (2)

H₂WO₄ → WO₃·0.33H₂O + H₂O (3)

Fig. 3 Schematic illustration of the proposed formation mechanism of the hierarchical flower-like WO₃·0.33H₂O.

As is known, the change of experiment parameters will greatly affect the morphology of the materials. In the present system, the addition of HCOOH plays an important role in the appearance of the final materials. Therefore, experiments with different volumes of HCOOH were carried out. Measurements of XRD and SEM were used to detect the samples and the corresponding results are shown in Fig. S2 and S3,† respectively. The diffraction peaks of the four samples are all in good accordance with the XRD pattern shown in Fig. 1a. When the volume of HCOOH used is 4 mL, the sample shows the morphology of a disc-like structure with a size of 2 μm (Fig. S3a†). However, the surfaces of the hierarchical flower-like structure become more smooth until the additive amount of HCOOH is 12 mL (Fig. S3b†). Detailed structure information (high magnification SEM images) and a simple structural representation are shown in Fig. 4.

Fig. 4 The detailed structure information (high magnification SEM images) and simple structural representations of hierarchical WO₃·0.33H₂O prepared with different amounts of HCOOH.

Through the above observation, we draw a conclusion that this phenomenon could be explained as follows: it is well-known that the ionization equilibrium of HCOOH (eqn (1)) can be strengthened by the increase of the concentration of HCOOH. The difference of the concentration of H⁺ in the reaction system will further affect the nucleation rate during the initial stage of nucleation, and then affect the growth process in reaching a minimum steady state of energy. Therefore, when the amount of HCOOH is small, the reversible reaction of eqn (1) takes place and creates a reaction environment with a low concentration of H⁺. Subsequently, the reaction described in eqn (2) occurs and new nuclei gradually assembled along the edge by oriented attachment, forming a disc-like structure. Although the extent of the degree of ionization could be largely accelerated, the concentration of HCOOH and H⁺ increases at the same time. As the amount of HCOOH is further increased to 12 mL, the initial nucleation and aggregation stages are the same as shown in Fig. 3. For the process of further growth, excessive HCOOH could adsorb on the surface of the WO₃·0.33H₂O nanosheets, forming a protective layer to prevent the deposition of generated WO₃·0.33H₂O tiny nanoparticles on the sheets, therefore, hierarchical flower-like WO₃·0.33H₂O with smooth petals are successfully developed.

According to observation from a TG curve, the calcination of hierarchical WO₃·0.33H₂O with the aforementioned three morphologies was carried out at 350 °C for 3 h in the air and they were labelled S-1, S-2 and S-3. The corresponding SEM images and XRD patterns are shown in Fig. 5. Obviously, the three samples retain the morphologies of the precursors well without deformation, as shown in Fig. 5a–c. All the diffraction peaks shown in Fig. 5d are perfectly indexed to WO₃ with a hexagonal phase (h-WO₃).

Fig. 5 SEM images of the as-prepared products prepared with different amounts of HCOOH after calcination at 350 °C for 3 h: (a) S-1, (b) S-2 and (c) S-3, and (d) the corresponding XRD pattern.

The electrochemical properties of electrode materials are largely dependent on morphology. In this work, the electrochemical performances of S-1, S-2 and S-3 are investigated systematically through testing these as anode electrodes for lithium ion batteries. This investigation was performed with a potential window ranging from 0.01 to 3 V. For the WO₃ electrode, the charge/discharge mechanism can be expressed as follows:^4,28

WO₃ + 6Li⁺ + 6e⁻ → W + 3Li₂O (4)

W + 3Li₂O → WO₃ + 6Li⁺ + 6e⁻ (5)

The theoretical capacity is calculated to be 693 mA h g⁻¹. The charge/discharge mechanism is consistent with the conversion reaction mechanism, which results in the charge/discharge capacities deriving from the reversible reaction of the formation of metal and lithium oxide. Fig. 6a shows the typical galvanostatic charge/discharge curves of S-2 measured at the current density of 100 mA g⁻¹. The characteristic plateau near 1 V of the curves followed by a sloped decrease in voltage during the first charging process indicates a conversion reaction mechanism.⁴ The observed overpotential between 1 and 1.6 V can be attributed to the kinetic nature of the proposed charging reaction of eqn (4).^29,30 Meanwhile for the discharge curves, there are no distinct voltage plateaus observed. Particularly, the electrochemical reaction between the electrode materials and lithium involves multiple steps for its decomposition and formation from the 1st discharge curve.³⁰ Compared with the initial discharge curve, there is a little difference for subsequent discharge curves, while the charge curves are identical to the first.

Fig. 6 (a) The typical charge/discharge curves and (b) CV curves of the S-2 electrode, (c) the first 100 cycle performances and (d) rate performances of the S-1, S-2 and S-3 electrodes.

Owing to the slow kinetic nature of the intrinsic properties expressed as eqn (4) and (5),^29,30 CV measurement is performed at a slow scan rate of 0.2 mV s⁻¹ with results shown in Fig. 6b. From the patterns it can be concluded that the first three CV curves well validate the observations from the charge/discharge curves. For the initial cycle curves, the peaks located at 0.3 V and 0.7 V correspond to the plateaus of the first discharge curve, which describe irreversible electrochemical processes. While for the second and third cycles, there is a very high overlap ratio, which indicates well the reversible reaction of the subsequent cycles.

The cycling and rate performances are two important indicators to assess an electrochemical device. Fig. 6c shows the cycling performances of S-1, S-2 and S-3 at the current density of 100 mA g⁻¹. Initially, the samples exhibit high discharge capacities of 2626.4, 2086.4, and 1686.1 mA h g⁻¹ in the first cycle for S-1, S-2 and S-3, respectively. However, there is a large capacity degradation for the subsequent cycles from the second cycle, which could be generally attributed to the irreversible electrochemical decomposition of the electrolyte, the electrode pulverization and the formation of a SEI layer on the surface of the electrode materials.^31,32 Interestingly, the discharge capacities remained at 377.2, 766.2, and 626 mA h g⁻¹ after 50 charge/discharge cycles. In particular, there is a small capacity decay after 100 cycles for S-2, which exhibits a discharge capacity of 720.5 mA h g⁻¹. Meanwhile, the coulombic efficiencies of the S-1, S-2 and S-3 electrodes in Fig. 6c are calculated to be constant with cycling, again demonstrating the excellent reversibility of the electrodes. The rate performance was also investigated and is shown in Fig. 6d. Obviously, S-2 shows remarkable discharge capacities, which are higher than S-1 and S-3 at the various current densities. The discharge capacity is 1284.5, 904.1, 829,3, 576.8 and 861 mA h g⁻¹ when the current density is 100, 200, 250, 500 and 100 mA g⁻¹, respectively. Furthermore, as the current density returns to 100 mA g⁻¹, the discharge capacity still remains at 787.9 mA h g⁻¹ after 80 discharge/charge cycles. These excellent electrochemical properties of S-2 are based on the hierarchical architecture which shortens the diffusion paths of the electrolyte ions, thus improving the dynamic performance, and the rough surface supplies more active sites.

The Nyquist plots of the S-1, S-2 and S-3 electrodes are shown in Fig. S4.† The electrochemical impedance spectra are composed of a depressed semicircle in the high frequency range and an angled straight line in the low frequency range.³³ The diameter of the semicircle represents the overlap between the SEI film and the interfacial charge transfer resistance, while the angled straight line is related to the lithium diffusion resistance. The curves were fitted with the equivalent circuit, which is shown in Fig. S4† and inset, the fitting results of the charge transfer resistance are 301.7 Ω, 205.9 Ω and 262.1 Ω for the S-1, S-2 and S-3 electrodes, respectively. These results indicate that the best electrochemical performance is of S-2, which corresponds to the high discharge capacities and long cycling stability.

4. Conclusions

In summary, h-WO₃ with three morphologies has been successfully prepared via simply adjusting the additive amount of HCOOH. The materials were investigated as electrode materials for lithium ion batteries. The hierarchical flower-like sample with a rough surface exhibits the highest discharge capacity and cycling stability, which indicates that the design and formation of hierarchical architectures is an effective method to improve electrochemical performances.

Acknowledgements

This work was supported by the National Natural of Science Foundation of China (Grant No. 21371101), 111 Project (B12015), MOE Innovation Team (IRT13022) of China and the Scientific Research Foundation of Qufu Normal University.

Notes and references

1. L. Liang, J. Zhang, Y. Zhou, J. Xie, X. Zhang, M. Guan, B. Pan and Y. Xie, Sci. Rep., 2013, 3, 1936.
2. Z.-G. Zhao and M. Miyauchi, Angew. Chem., 2008, 120, 7159–7163.
3. J. Qin, M. Cao, N. Li and C. Hu, J. Mater. Chem., 2011, 21, 17167–17174.
4. S. Yoon, C. Jo, S. Y. Noh, C. W. Lee, J. H. Song and J. Lee, Phys. Chem. Chem. Phys., 2011, 13, 11060–11066.
5. K.-H. Chang, C.-C. Hu, C.-M. Huang, Y.-L. Liu and C.-I. Chang, J. Power Sources, 2011, 196, 2387–2392.
6. L. Gao, X. Wang, Z. Xie, W. Song, L. Wang, X. Wu, F. Qu, D. Chena and G. Shen, J. Mater. Chem. A, 2013, 1, 7167–7173.
7. Y.-H. Wang, C.-C. Wang, W.-Y. Cheng and S.-Y. Lu, Carbon, 2014, 69, 287–293.
8. X. Xiao, T. Ding, L. Yuan, Y. Shen, Q. Zhong, X. Zhang, Y. Cao, B. Hu, T. Zhai, L. Gong, J. Chen, Y. Tong, J. Zhou and Z. L. Wang, Adv. Energy Mater., 2012, 2, 1328–1332.
9. H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell and K. Kalantar-zadeh, Adv. Funct. Mater., 2011, 21, 2175–2196.
10. J. Ma, J. Zhang, S. Wang, T. Wang, J. Lian, X. Duan and W. Zheng, J. Phys. Chem. C, 2011, 115, 18157–18163.
11. K. Kalantar-zadeh, A. Vijayaraghavan, M.-H. Ham, H. Zheng, M. Breedon and M. S. Strano, Chem. Mater., 2010, 22, 5660–5666.
12. D. Chen, L. Gao, A. Yasumori, K. Kuroda and Y. Sugahara, Small, 2008, 4, 813–1822.
13. Y.-L. Ma, L. Zhang, X.-F. Cao, X.-T. Chen and Z.-L. Xue, CrystEngComm, 2010, 12, 1153–1158.
14. J. Yang, L. Jiao, Q. Zhao, Q. Wang, H. Gao, Q. Huan, W. Zheng, Y. Wang and H. Yuan, J. Mater. Chem., 2012, 22, 3699–3701.
15. Z.-G. Zhao and M. Miyauchi, J. Phys. Chem. C, 2009, 113, 6539–6546.
16. J. Shi, G. Hu, R. Cong, H. Bu and N. Dai, New J. Chem., 2013, 37, 1538–1544.
17. L. Zhou, J. Zou, M. Yu, P. Lu, J. Wei, Y. Qian, Y. Wang and C. Yu, Cryst. Growth Des., 2008, 8, 3993–3998.
18. J. Li, J. Huang, C. Yu, J. Wu, L. Cao and K. Yanagisawa, Chem. Lett., 2011, 40, 579–581.
19. J. Wang, P. S. Lee and J. Ma, Cryst. Growth Des., 2009, 9, 2293–2299.
20. Z. Gu, T. Zhai, B. Gao, X. Sheng, Y. Wang, H. Fu, Y. Ma and J. Yao, J. Phys. Chem. B, 2006, 110, 23829–23836.
21. Z. Gu, Y. Ma, W. Yang, G. Zhang and J. Yao, Chem. Commun., 2005, 3597–3599.
22. C. Ng, C. Ye, Y. H. Ng and R. Amal, Cryst. Growth Des., 2010, 10, 3794–3801.
23. S. Balaji, Y. Djaoued, A. Albert, R. Z. Ferguson and R. Brüning, Chem. Mater., 2009, 21, 1381–1389.
24. Z. Gu, H. Li, T. Zhai, W. Yang, Y. Xia, Y. Ma and J. Yao, J. Solid State Chem., 2007, 180, 98–105.
25. Y. Liu, Y. Jiao, Z. Zhang, F. Qu, A. Umar and X. Wu, ACS Appl. Mater. Interfaces, 2014, 6, 2174–2184.
26. D. Li, M. H. Nielsen, J. R. I. Lee, C. Frandsen, J. F. Banfield and J. J. D. Yoreol, Science, 2012, 336, 1014–1018.
27. Z. Xing, B. Geng, X. Li, H. Jiang, C. Feng and T. Ge, CrystEngComm, 2011, 13, 2137–2142.
28. X. Duan, S. Xiao, L. Wang, H. Huang, Y. Liu, Q. Li and T. Wang, Nanoscale, 2015, 7, 2230–2234.
29. N. Kumagai, Y. Umetzu, K. Tanno and J. Pereira-Ramos, Solid State Ionics, 1996, 86, 1443–1449.
30. W. J. Li and Z. W. Fu, Appl. Surf. Sci., 2010, 256, 2447–2452.
31. Q. Su, D. Xie, J. Zhang, G. Du and B. Xu, ACS Nano, 2013, 7, 9115–9121.
32. D. Larcher, C. Masquelier, D. Bonnin, Y. Chabre, V. Masson, J. B. Leriche and J. M. Tarascon, J. Electrochem. Soc., 2003, 150, A133–A139.
33. Y. Sun, X. Hu, W. Luo and Y. Huang, ACS Nano, 2011, 5, 7100–7107.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1–S4. See DOI: 10.1039/c5ra26645a

---