Synthesis of novel morphologies of Li₂FeSiO₄/C micro/nano composites by a facile hydrothermal method

Abstract

Carambola and jujube-seed-shaped Li₂FeSiO₄ assembled by nanoplates have been successfully synthesized by a simple hydrothermal method. The different morphologies are induced by two different iron precursors, which affect the self-organization behaviour of primary particles of Li₂FeSiO₄. The electrochemical performances of the two Li₂FeSiO₄/C composites are also influenced by the different morphology.

---

The increasing demand for high-power and high-energy batteries has motivated continuous research on lithium-ion batteries (LIBs). Recently, Li₂FeSiO₄ as one example of a polyanion-type cathode has attracted considerable interest due to its low cost, environmentally friendly nature, and high theoretical capacity (166 mA h g⁻¹ for one Li⁺ ion and 332 mA h g⁻¹ for two Li⁺ ions exchange).^1–4 However, the low electric conductivity and slow lithium ion diffusion of Li₂FeSiO₄ limit its high-rate performance.^2,5 Therefore, various methods, including conductive carbon coating,⁶ particle size reduction⁷ and metallic ion doping,⁸ are adopted to make up for these shortages.

In addition to the abovementioned methods, morphology control is an alternative method to achieve superior electrochemical properties, which has been demonstrated in many cathode materials.^9–12 Various nanostructures of Li₂FeSiO₄ such as sphere-like particles or one-dimensional structures have been investigated.^13–15 For example, Wu et al. synthesized nanoworm-like Li₂FeSiO₄/C composites using amphiphilic triblock copolymer P123 as the structure-directing agent.¹⁶ Similarly, sphere-like Li₂FeSiO₄ particles with ∼50 nm size were synthesized via a modified sol-gel method.¹⁵ However, although nanostructured electrode materials bring high specific capacity, they adversely affect the tap density and volumetric energy density.¹⁷ On the basis of the investigations of LiFePO₄/C composites, it is concluded that hierarchical morphologies assembled by nanoscale primary particles could bring not only high reversible capacity but also good rate capability.^10,12,17,18 In light of this, the assembly of nanoscale building blocks into hierarchical microstructures might be a good solution. Recently, Yang and his coworkers have synthesized hierarchical shuttle-like Li₂FeSiO₄ for the first time by a hydrothermal reaction.¹⁹ However, their final product was not pure (Fe₃O₄ as the impurity) and their synthetic method relied on a long reaction time (8 days). Hence, it is necessary to develop a simple and rapid method to synthesize a micro/nano structure of Li₂FeSiO₄. On the other hand, most hierarchical structures of electrode materials were induced by templating methods,^20,21 while the influence of raw materials such as iron precursors on the particle geometries of Li₂FeSiO₄ has not been investigated to the best of our knowledge.

Herein, we show that different iron precursors have a profound effect on the particle morphology of Li₂FeSiO₄. Carambola and jujube-like Li₂FeSiO₄ assembled by a large number of nanoplates are obtained by a facile hydrothermal reaction at 200 °C for only 12 h. XRD patterns and TEM images are applied to follow the evolution of the product. Moreover, Li₂FeSiO₄/C was obtained via ball-milling beta-cyclodextrin, which acted as a superior carbon source in our previous work²² with Li₂FeSiO₄ and was then sintered under an Ar (95%)/H₂ (5%) atmosphere.

The XRD patterns of LFS-1/C (using (NH₄)₂Fe(SO₄)₂·6H₂O as iron precursor) and LFS-2/C (using FeSO₄·7H₂O as iron precursor) are shown in Fig. 1. All the peaks in the XRD patterns could be indexed to an orthorhombic structure with space group Pmn2₁, which is in good agreement with the previous reports.^23–26 The absence of impurity phases, such as iron oxides or lithium silicate phases, in the final products indicates the high purity of LFS-1/C and LFS-2/C.

Fig. 1 The XRD patterns of (A) LFS-1/C and (B) LFS-2/C composites; inset is the crystal structure of Li₂FeSiO₄ viewed along the c-axis.

The morphologies of LFS-1/C and LFS-2/C are investigated by SEM, TEM and HRTEM imaging. It can be seen from Fig. 2A that LFS-1/C exhibits a carambola shape with a size of ∼2 μm in length and ∼0.5 μm in width. The carambola consists of nanoplates and the nanoplates are around 180 nm in width and ∼40 nm in thickness. More interestingly, the carambola is a hollow structure, as shown in Fig. 2B. It should be pointed out that the morphology and size of pristine LFS-1 remain basically unchanged after coating with carbon (Fig. S1a and b†). This demonstrates that the presence of carbon could play an important role in preventing the growth of LFS-1 during sintering. To the best of our knowledge, this novel morphology of Li₂FeSiO₄ has been synthesized for the first time. Such a well-organized structure is expected to facilitate electrolyte penetration into the electrode particles by providing more interfacial areas between the electrode material and the electrolyte.¹⁹ As for LFS-2/C, the SEM and TEM images (Fig. 2C and D) show a solid jujube-seed shape with a size of ∼1.5 μm in length and ∼0.5 μm in width. It is clear that the jujube-seed particles consist of compact nanoparticles and nanoplates with irregular shape. Similarly, the morphology and size of LFS-2 are retained after carbon coating (Fig. S1c and d†). Energy dispersive X-ray (EDX) element maps of the Si, Fe and O signals of the LFS-1/C and LFS-2/C composites indicate that the Fe, O and Si elements are well distributed for both samples (Fig. S2 and S3†). High-resolution TEM (HRTEM) images of the outer edges of the framework are shown in Fig. 2E and F. Crystal lattice stripes of the two samples are observed with d-spacing of 3.41 Å and 3.90 Å, which correspond to the (002) and (101) planes of orthorhombic Li₂FeSiO₄ crystals. As expected, a continuous and uniform layer of carbon is covered on the surfaces of the LFS-1/C and LFS-2/C composites utilizing beta-cyclodextrin as a carbon source²² and the thicknesses of the carbon layers are only around 2 nm and 6 nm, respectively (Fig. 2E and F). The Raman spectra of the as-synthesized LFS-1/C and LFS-2/C are shown in Fig. S4A and B.† For both samples, the peaks at 1320 (peak 2) and 1590 cm⁻¹ (peak 4) are the D band and G band of sp²-type carbon, while the peaks at around 1180 cm⁻¹ (peak 1) and 1460 cm⁻¹ (peak 3) are related to sp³-type carbon.^2,16 The I_D/I_G ratios (intensity ratios of the D and G bands) of LFS-1/C and LFS-2/C are fitted to 0.96 and 0.95, respectively, which can be used to evaluate the graphitization degree of the composite.^2,22 The ratios of I_D/I_G for the two samples are quite close, indicating a similar degree of graphitization. By TG analysis, the mass fraction of carbon for the LFS-1/C and LFS-2/C composites is determined to be about 3.72 wt% and 4.26 wt%, respectively (Fig. S4C and D†). In addition, the specific surface areas of the two samples are investigated via nitrogen adsorption–desorption measurements. As a result of its hollow carambola-shaped structure, the measured Brunauer–Emmett–Teller (BET) area for LFS-1/C is 36.4 m² g⁻¹, which is higher than that of LFS-2/C (16.3 m² g⁻¹) as shown in Fig. S5.†

Fig. 2 The SEM and TEM images of sample LFS-1/C (A and B, respectively), and sample LFS-2/C (C and D, respectively), and the HRTEM images of LFS-1/C (E) and LFS-2/C (F).

To confirm the possible formation mechanism of carambola LFS-1 and jujube-seed LFS-2, intermediate particles were collected at different reaction times. We take LFS-1 as an example to investigate the possible growth process. As shown in Fig. 3A, XRD patterns of all intermediate products are presented. As for LFS-1, the product is mainly composed of an amorphous substance after 4 h reaction. The peaks of the XRD patterns correspond to Li₂SiO₃ and Fe₃O₄ after aging for up to 6 h. When the reaction time increases to 8 h, the product becomes Li₂FeSiO₄ with impurities. After 12 h, a pure phase of Li₂FeSiO₄ is observed. A possible formation mechanism of LFS-1 can be proposed as follows: first, Li₂SiO₃ forms by reaction of Li⁺, OH⁻ and SiO₂:

2Li⁺ + 2OH⁻ + SiO₂ → Li₂SiO₃ + H₂O (1)

Fig. 3 The XRD patterns (A) and TEM images (B) of the as-prepared LFS-1.

Moreover, Fe₃O₄ has also been formed resulting from Fe²⁺ reacting with trace O₂ under OH⁻ and ethylene glycol:

6Fe²⁺ + 12OH⁻ + O₂ → 2Fe₃O₄ + 6H₂O (2)

As the reaction time increases, Li₂FeSiO₄ forms by reaction:

The LFS-2 sample experiences similar stages except that the formation of pure Li₂FeSiO₄ is much faster than with the LFS-1 sample (Fig. S6A†). It is suggested that the formation phase for Li₂FeSiO₄ should be related to the reaction conditions such as precursors or solvents. For example, Fe(Ac)₂·4H₂O was chosen as the iron resource in Yang' work,¹⁹ which may have contributed to the long reaction time of 8 days to obtain Li₂FeSiO₄ in their system.

The morphology evolution for LFS-1 is shown in Fig. 3B. It is found that needle-like particles appear under hydrothermal conditions after 2 h, and then gather and form isosceles triangle-like particles after 3.5 h. After 4.5 h, a hollow shuttle-like morphology is observed and this grows very regular and uniform under hydrothermal conditions after 6 h. Then, these hollow shuttle-like particles aggregate after 8 h. Finally, pure carambola LFS-1 can be formed when the reaction time reaches 12 h. For LFS-2, seaweed-like particles appear under hydrothermal conditions after 0.5 h, then aggregate into irregular bulk-like particles and finally form a shuttle-like morphology only after 2.5 h. (Fig. S6B†). The difference in the precursor solutions of LFS-1 and LFS-2 is the iron source, i.e., (NH₄)₂Fe(SO₄)₂ for LFS-1 and FeSO₄ for LFS-2. Based on Lu's work,²⁷ we speculate that NH₄⁺ ions can easily adhere to one or more certain surfaces of Li₂FeSiO₄ particles and this helps etch (dissolve) cations from the surfaces of the Li₂FeSiO₄ particles, which allows the morphology to continuously evolve and results in carambola LFS-1 with a hollow structure under hydrothermal conditions. However, for LFS-2, no hollow structures were found due to the absence of NH₄⁺. In order to understand the effect of temperature on the structure and morphology of Li₂FeSiO₄ using the two different iron sources, we also synthesized Li₂FeSiO₄ under the same experimental conditions except that the reaction temperatures are 160 °C and 180 °C (Fig. S7†). For “LFS-1” with reaction temperature 160 °C as shown in Fig.S7A,† the diffraction patterns can be indexed to Fe₂SiO₄ (JCPDS Card no. 34-0178). Note that the main peaks for the sample at 180 °C are assigned to Li₂FeSiO₄, but Fe₂SiO₄ as an impurity phase is present in the diffraction pattern. Similarly, for “LFS-2” with reaction temperature at 160 °C or 180 °C, Li₂FeSiO₄ cannot be obtained or the impurity phase emerges (Fig. S7D†). This indicated that Li₂FeSiO₄ cannot be formed or the phase is not pure when the reaction temperature is lower than 200 °C in our experiment. From the TEM images of LFS at 160 °C and 180 °C (Fig. S7B, C, E and F†), it is observed that the morphologies of the four samples are similar to those of LFS at 200 °C.

On the basis of the above experimental evidence, a possible schematic illustration for the formation of carambola LFS-1 is given in Fig. 4. The process can be classified into three steps, the first of which is nucleation and growth. Secondly, the precursor nuclei quickly grow into primary nanoplates. In the last step, the primary particles continuously self-assemble into a micro-carambola morphology, driven by the reduction in the surface free energy.²⁸ However, to further understand the detailed information of the effect of NH₄⁺ on the hollow carambola morphology, more work needs to be done. Nevertheless, the two novel morphologies of Li₂FeSiO₄ in our case are obtained just by choosing different iron resources, which are inexpensive and environmentally friendly. We believe that this simple, convenient and rapid preparation route is suitable to commercialize, compared to other synthesis methods such as the spray pyrolysis²⁹ and sol–gel method.³⁰

Fig. 4 The schematic illustration of the formation of LFS-1 hierarchical structure.

As a cathode material in a Li-ion battery, we have also compared the electrochemical performances of carambola LFS-1 and jujube-seed LFS-2. As presented in Fig. S8A and B,† the carambola LFS-1 sample shows better electrochemical properties, including reversible capacity for the initial cycle and cyclability compared to those of the jujube-seed LFS-2 sample. In order to understand the effect of morphology on the electrochemical performance of Li₂FeSiO₄ and the corresponding dynamic behaviour, we also investigated the direct current resistance of LFS/C samples (Fig. S8C and D†). We also assessed the tap densities of the carambola, jujube-seed and nanoparticle (synthesized by a solid-state method) Li₂FeSiO₄, which are 1.2 g cm⁻³, 1.1 g cm⁻³ and 1.0 g cm⁻³, respectively. By calculation, the volumetric energy densities are 495, 341.8 and 302.5 W h L⁻¹, respectively. Therefore, the hierarchical microstructure in our case can improve the tap density and volumetric energy density for a Li₂FeSiO₄ electrode material. From these experimental results, it is suggested that the carambola LFS-1/C possesses several advantages, resulting in good electrochemical performance: (1) the morphology with an outer open and inner hollow structure reduces the lithium diffusion length, avoiding loss of electrochemical activity of the electrode material, especially the inner materials, during cycles; (2) a thinner carbon layer than LFS-2/C not only improves the electronic conductivity, but also reduces the hindrance to lithium ion diffusion; and (3) a larger specific surface area enhances the contact area between the electrolyte and active particles, indicating more active sites during the electrochemical reaction procedure.

In conclusion, we have successfully synthesized carambola and jujube-seed Li₂FeSiO₄/C by a facile hydrothermal method. The different iron precursors have a profound effect on the particle morphology of Li₂FeSiO₄ and the corresponding electrochemical performance. In addition, the synthetic strategy described in our work for making different morphologies and useful structures may be extended to fabricate other types of multifunctional materials for energy storage, catalysis and other applications.

Acknowledgements

This work is partially supported by the 973 Project of China (no. 2011CB935901), the National Natural Science Fund of China (nos 21203111, 91022033), the Shandong Provincial Natural Science Foundation for Distinguished Young Scholar (JQ201205), the Independent Innovation Foundation of Shandong University (no. 2012 ZD007), and the start-up funding for new faculties in Shandong University.

Notes and references

1. Z. L. Gong and Y. Yang, Energy Environ. Sci., 2011, 4, 3223.
2. Z. M. Zheng, Y. Wang, A. Zhang, T. Zhang, F. Cheng, Z. Tao and J. Chen, J. Power Sources, 2012, 198, 229.
3. M. Bini, S. Ferrari, D. Capsoni, C. Spreafico, C. Tealdi and P. Mustarelli, J. Solid State Chem., 2013, 200, 70.
4. Z. X. Chen, S. Qiu, Y. L. Cao, J. F. Qian, X. P. Ai, K. Xie, X. B. Hong and H. X. Yang, J. Mater. Chem. A, 2013, 1, 4988.
5. X. Y. Fan, Y. Li, J. J. Wang, L. Gou, P. Zhao, D. L. Li, L. Huang and S. G. Sun, J. Alloys Compd., 2010, 493, 77.
6. J. L. Yang, X. C. Kang, L. Hu, X. Gong and S. C. Mu, J. Mater. Chem. A, 2014, 2, 6870.
7. D. Rangappa, K. D. Murukanahally, T. Tomai, A. Unemoto and I. Honma, Nano Lett., 2012, 12, 1146.
8. Y. S. Li, X. Cheng and Y. Zhang, J. Electrochem. Soc., 2012, 159, A69.
9. F. Wang, J. Yang, P. F. Gao, Y. N. NuLi and J. L. Wang, J. Power Sources, 2011, 196, 10258.
10. J. F. Qian, M. Zhou, Y. L. Cao, X. P. Ai and H. X. Yang, J. Phys. Chem. C, 2010, 114, 3477.
11. F. Jiao, J. Bao, A. H. Hill and P. G. Bruce, Angew. Chem., Int. Ed., 2008, 27, 9711.
12. K. Saravanan, M. V. Reddy, P. Balaya, H. Gong, B. V. R. Chowdari and J. J. Vittal, J. Mater. Chem., 2009, 19, 605.
13. G. Peng, L. L. Zhang, X. L. Yang, S. Duan, G. Liang and Y. H. Huang, J. Alloys Compd., 2013, 570, 1.
14. J. L. Yang, X. C. Kang, L. Hu, X. Gong, D. P. He, T. Peng and S. C. Mu, J. Alloys Compd., 2013, 572, 158.
15. H. Zhu, X. Z. Wu, L. Zan and Y. X. Zhang, Electrochim. Acta, 2014, 117, 34.
16. X. Z. Wu, X. M. Wang and Y. X. Zhang, ACS Appl. Mater. Interfaces, 2013, 5, 2510.
17. C. W. Sun, S. Rajasekhara, J. B. Goodenough and F. Zhou, J. Am. Chem. Soc., 2011, 133, 2132.
18. M. Wang, Y. Yang and Y. X. Zhang, Nanoscale, 2011, 3, 4434.
19. J. L. Yang, X. C. Kang, D. P. He, T. Peng, L. Hu and S. C. Mu, J. Power Sources, 2013, 242, 171.
20. S. J. Kim, J. Suk, Y. J. Yun, H. K. Jung and S. Choi, Phys. Chem. Chem. Phys., 2014, 16, 2085.
21. X. Z. Kong, T. Mei, Z. Xing, N. Li, Z. Q. Yuan, Y. C. Zhu and Y. T. Qian, Int. J. Electrochem. Sci., 2012, 7, 5565.
22. L. E. Li, J. Liu, L. Chen, H. Y. Xu, J. Yang and Y. T. Qian, RSC Adv., 2013, 3, 6847.
23. Z. L. Gong, Y. X. Li, G. N. He, J. Li and Y. Yang, Electrochem. Solid-State Lett., 2008, 11, A60.
24. L. Qu, S. H. Fang, L. Yang and S. I. Hirano, J. Power Sources, 2012, 217, 243.
25. P. J. Zuo, T. Wang, G. Y. Cheng, X. Q. Cheng, C. Y. Du and G. P. Yin, RSC Adv., 2012, 2, 6994.
26. X. M. Wang, C. X. Qing, Q. T. Zhang, W. F. Fan, X. B. Huang, B. P. Yang and J. F. Cui, Electrochim. Acta, 2014, 134, 371.
27. Z. G. Lu, H. L. Chen, R. Robert, B. Y. X. Zhu, J. Q. Deng, L. J. Wu, C. Y. Chung and C. P. Grey, Chem. Mater., 2011, 23, 2848.
28. H. Yang, X. L. Wu, M. H. Cao and Y. G. Guo, J. Phys. Chem. C, 2009, 113, 3345.
29. B. Shao, Y. Abe and I. Taniguchi, Powder Technol., 2013, 235, 1.
30. C. Deng, S. Zhang, G. S. Zhao, Z. Dong, Y. Shang, Y. X. Wu and B. D. Zhao, J. Electrochem. Soc., 2013, 160, A1457.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, SEM and TEM images of pristine LFS-1 and LFS-2, EDX element maps of Fe, Si and O in sample LFS-1/C, EDX element maps of Fe, Si and O in sample LFS-2/C, Raman spectra and TGA curves of LFS-1/C and LFS-2/C, nitrogen adsorption–desorption isotherms of LFS-1/C and LFS-2/C, XRD patterns and TEM images of the as-prepared LFS-2, XRD patterns of (A) LFS-1/C and (D) LFS-2/C composites at 160 °C and 180 °C; TEM images of sample LFS-1/C at 160 °C (B) and 180 °C (C), and sample LFS-2/C at 160 °C (E) and 180 °C (F), electrochemical performances of two samples: the typical charge and discharge curves of LFS-1/C and LFS-2/C; the plots of the direct current resistances of LFS-1/C and LFS-2/C vs. state of charge and depth of discharge after 50 cycles. See DOI: 10.1039/c4ra06392a

---