Multi-shelled LiMn₂O₄ hollow microspheres as superior cathode materials for lithium-ion batteries

Abstract

Owing to its environmental-benignity, low-cost and abundance, spinel LiMn₂O₄ has long been considered as a promising cathode material for lithium-ion batteries (LIBs). However, the low electronic conductivity, small lithium diffusion coefficient and poor capacity retention hindered its further development and application. Herein, we report the synthesis of multi-shelled LiMn₂O₄ hollow microspheres through a hard template method, with the composition, shell number, shell thickness and porosity accurately controlled. Benefitting from the structural superiorities of multi-shelled hollow structures, the triple-shelled LiMn₂O₄ hollow microsphere exhibits a better cycling stability than all the reported results based on un-coated or un-doped LiMn₂O₄ (the capacity fading rate is 0.10% per cycle).

---

Introduction

Lithium-ion batteries (LIBs), owing to their remarkable advantages such as light-weight, no memory effect, and environmental friendliness, have been considered as the most promising candidate to meet the energy and environmental requirements in the 21^st century.^1,2 The commercially available cathode materials are mostly Li_xCoO₂ with a specific capacity of 140 mA h g⁻¹ based on a practical value of x from 1 to 0.5. However, the toxicity, high-cost and scarcity of Co limit its large-scale application in LIBs.³ As a result, it's urgently needed to develop other cathode materials to meet the further development of next-generation LIBs. Spinel-structured lithium manganese oxide (LiMn₂O₄), with a high potential of 4 V and a theoretical specific capacity of 148 mA h g⁻¹, has long been studied as an attractive cathode material for high-power LIBs due to the non-toxicity, low-cost and abundance of Mn.^4–8 Unfortunately, the practical application of LiMn₂O₄ has been impeded by its kinetic problems such as low electronic conductivity and small lithium diffusion coefficient, also poor capacity retention caused by Mn dissolution in the electrolyte.^9–12 One effective strategy to overcome these drawbacks is building hollow micro/nanostructured materials to replace the bulk one as cathodes of LIBs. The advantages of adopting hollow micro/nanostructures are obvious and listed below:^13–15 (1) the enhanced surface-to-volume ratio can provide more effective specific surface area, thus a higher specific capacity; (2) the improved activity and shortened transport length for both ions and electrons lead to an improved rate capability and power density; (3) more importantly, with the multi-shelled hollow structures, different shells can support each other and the exterior shell protects the interior shells from dissolution, leading to an improved structural stability and better cycling performance.

Given the above considerations, great efforts have been devoted to the synthesis of hollow micro/nanostructured cathode materials, such as hollow spheres,^16–18 hollow octahedra,¹⁹ hollow spindles²⁰ and hollow boxes.²¹ However, all the achieved hollow structures are only single-shelled or single-shelled dominated in the final products.^22,23 Recently, though some other metal oxide multi-shelled hollow microspheres have been synthesized using carbonaceous microspheres (CMSs) as templates,^24–28 most of the products contain a single metal. There are a few products containing two kinds of metal elements, yet only restricted to metal elements with similar qualities. Due to the distinct nature of Li and Mn elements, the synthesis of multi-shelled LiMn₂O₄ hollow microspheres (MS-LiMn₂O₄-HMSs) still remains rather challenging. Herein, we report an effective approach to produce MS-LiMn₂O₄-HMSs with the composition, shell number (triple or quadruple), shell thickness and porosity accurately controlled. Profiting from the superiorities of multi-shelled hollow nanostructures, MS-LiMn₂O₄-HMSs as cathode materials for LIBs exhibit highly improved lithium-storage performance: the triple-shelled one exhibits a better cycling stability than all the reported results based on un-coated or un-doped LiMn₂O₄ (the capacity fading rate is only 0.10% per cycle even to 200 cycles) (see Table S1†).

Results and discussion

As Fig. S1† shows, the compositions of the final products are quite different at different ratios of the Li precursor (LiAc) to Mn precursor (Mn(NO₃)₂·4H₂O). Only when the adding ratio of the Li precursor to Mn precursor is around 10 : 1 can LiMn₂O₄ products be achieved (Fig. 1e). It's also worth mentioning that the ratio of Li to Mn in the precursor solution is much higher than that in the final products. The reason may be that CMSs are more inclined to adsorb Mn(II) ions compared to Li (I) ions, since the former has more charges than the latter, leading to a stronger electrostatic interaction between Mn(II) ions and negatively-charged CMS templates.²⁹ Besides, the empty d orbitals of Mn²⁺ are well ready to accept the p electrons donated by oxygen (O) in the CMSs, making strong coordination interaction between Mn(II) ions and CMSs, thus Mn(II) ions can stay stable in CMSs. Moreover, more metal precursor will be adsorbed at a larger concentration. Therefore, to guarantee that enough Li is adsorbed by CMSs, the concentration of the Li precursor should be much higher than that of the Mn precursor.

Fig. 1 (a, b) TEM and (c, d) SEM images of LiMn₂O₄ multi-shelled hollow microspheres: (a, c) triple-shelled; (b, d) quadruple-shelled. (e) XRD patterns of as-prepared LiMn₂O₄ products: i and ii indicate triple- and quadruple-shelled hollow microspheres, respectively. (f) HRTEM with inserted SAED characterization of the triple-shelled hollow microspheres.

The annealing process is also a key parameter which controls the morphology and structure of the final microsphere products, wherein the heating rate is one of the most important factors which can cause great discrepancy between the rate of metal oxide shell formation (v₁) and CMS contraction (v₂). Normally, the slow heating rate can benefit the shell number. Under a slow heating rate (1 °C min⁻¹), both the speed of oxide-crystallization and the difference between v₁ and v₂ decrease. And the outer oxide will shrink along with the shrinking of the inner CMS core during the early period of annealing as there is no enough oxide to sustain the formation of the oxide shell, thus accumulating to form a thick shell. Then the thick shell further shrinks and separates to double shells, and finally results in triple-shelled hollow microspheres with relatively small diameters (3S-HMSs) (Fig. 1a and c). By contrast, bigger double-shelled hollow microspheres (2S-HMSs) were obtained at a higher heating rate (10 °C min⁻¹) (Fig. S2†) (yet the phase is not pure LiMn₂O₄).

Besides the heating rate, the adsorption duration also matters a lot in controlling the morphology and structure of the products. It's easy to understand that the adsorption amount and the penetration depth into the CMSs increase along with the increase of the adsorption duration. As a result, by prolonging the adsorption duration and further slowing the heating rate (0.5 °C min⁻¹), quadruple-shelled hollow microspheres (4S-HMSs) are obtained (Fig. 1b and d) (see Experimental details in Table S2†).

Through the above methods, triple- and quadruple-shelled LiMn₂O₄ hollow microspheres with high purity were prepared. From the transmission electron microscopy (TEM) (Fig. 1a and b) and scanning electron microscopy (SEM) (Fig. 1c and d) characterization, we can see that the size is uniform, with diameters of about 0.9 and 1.2 μm for 3S- and 4S-LiMn₂O₄-HMSs respectively. The X-ray diffraction (XRD) patterns (Fig. 1e) show that both the products correspond to spinel type LiMn₂O₄ (JCPDS card no. 35-0782). Besides, the peaks are rather sharp and no additional impurity peaks were observed, indicating good crystallinity and purity of the products.^30,31 From the high resolution TEM (HRTEM) image (Fig. 1f), we can also see that the product is highly crystallized, with a d-space of 0.48 nm commonly observed, corresponding to the (111) crystal facet of LiMn₂O₄. Considering the weight loss stopped at about 348 °C according to TGA-DTA (thermogravimetric analysis–differential thermal analysis) data (Fig. S3†), the excellent crystallinity of the products can be ascribed to the optimal annealing treatment conditions (600 °C, 2 h) which guarantee the complete combustion of CMS templates and crystallization of LiMn₂O₄. To further test the chemical composition of the as-prepared MS-LiMn₂O₄-HMSs, X-ray photoelectron spectroscopy (XPS) was carried out (Fig. S4†). The two peaks located at 641.1 eV and 642.55 eV correspond to Mn³⁺, while the other peak located at 643.1 eV corresponds to Mn⁴⁺.¹²

The continuous shells are composed of small nanoparticles according to the structural observation of the typical 3S-LiMn₂O₄-HMS (Fig. S5†). The nitrogen adsorption–desorption analysis (Fig. S6†) shows that the specific surface area of 4S-LiMn₂O₄-HMS is larger than that of the triple-shelled (47.04 vs. 31.69 m² g⁻¹) (Table S3†). Fig. S7† demonstrates that nanopores of 2.5 and 4 nm mainly exist in the triple- and quadruple-shelled hollow microspheres, respectively. The electrolyte can directly penetrate into the interior of the hollow microspheres through these pores, thus shortening the transfer path and improving the kinetic characters.

The cyclic voltammograms (CV) curves of the MS-LiMn₂O₄-HMSs samples at a scan rate of 1 mV s⁻¹ in a potential range from 3.5 to 4.3 V are shown in Fig. 2a. Two typical redox peaks around 3.95/4.05 V and 4.1/4.2 V (vs. Li⁺/Li) are clearly observed for all the LiMn₂O₄ products. Two anode peaks at about 4.05 and 4.2 V are corresponding to Li-extraction from the crystal lattice and the formation of Li_0.5Mn₂O₄ and λ-MnO₂ respectively, while the two cathode peaks at around 3.95 and 4.1 V indicate the reversible Li-insertion process, which is similar to the previously reported results.^32,33 Moreover, with the same loading amount, the current density of 4S-LiMn₂O₄-HMSs is larger than that of 3S-LiMn₂O₄-HMSs, demonstrating a faster charge separation and better electron conductivity. The Nyquist plots in Fig. 2b also support this hypothesis. Both the samples display a semicircle in the high frequency region and a sloping straight line in the low frequency range, presenting the charge transfer resistance (R_ct) and diffusion of lithium in the solid (Z_w), respectively.³⁴ The smaller semicircle diameter for the 4S-HMS based electrode shows a smaller R_ct compared to the 3S-HMS based electrode, indicating quicker charge transfer. This discrepancy may be ascribed to the thinner shell and larger surface area of 4S-LiMn₂O₄-HMSs, which can enhance the contact area of electrolyte–electrode, and facilitate the electrolyte transport and lithium ion diffusion. By contrast, limited by the thick shell and small specific surface area, the triple-shelled one shows the smaller charge transfer resistance.

Fig. 2 (a) Cyclic voltammetry (CV) curves and (b) electrochemical impedance spectroscopy (EIS) for multi-shelled LiMn₂O₄ hollow microspheres with the simplified Randles equivalent circuit model inset (R_Ω, external resistance; R_ct, charge transfer resistance; CPE1, constant phase element; Z_w, Warburg impedance).

The lithium-storage properties of the MS-LiMn₂O₄-HMSs as cathode materials for LIBs are tested by fabricating a standard LiMn₂O₄/Li half-cell configuration. Taking 3S-LiMn₂O₄-HMS as an example, two plateaus located at about 4.10 and 3.95 V in the discharge curves, and another two plateaus at around 4.05 and 4.2 V in the charge curves are obviously noticed (Fig. 3a and S8†), which is consistent with the above CV results. Moreover, the plateaus can still be observed even after 100 cycles, indicating good reversibility of the 3S-LiMn₂O₄-HMSs.

Fig. 3 (a) Discharge–charge curves of the triple-shelled hollow microspheres at different cycle numbers with a constant rate of 1 C (assuming 1 C = 148 mA g⁻¹), (b) charge and discharge capacities and coulombic efficiencies (CEs) of multi-shelled hollow spheres at a rate of 1 C, and (c) rate capability test of the triple-shelled hollow spheres according to the cycling rate sequence: 1C, 2 C, 4 C, 8 C, 10 C, 1 C. All tests were performed between 3.5 and 4.3 V.

The specific capacity and cycling performance of LiMn₂O₄ hollow microspheres with different shell numbers are compared in Fig. 3b. As expected by CV and EIS results, the 4S-LiMn₂O₄-HMSs exhibit a higher initial discharge specific capacity as high as 143.4 mA h g⁻¹, while that of the 3S-HMSs is 115.5 mA h g⁻¹. The reason for the specific capacity of the 4S-LiMn₂O₄-HMSs being superior to that of the 3S-LiMn₂O₄-HMSs may be the fact that a larger specific surface area of the 4S-LiMn₂O₄-HMSs can provide more surface lithium-storage sites, a larger electrode/electrolyte contact area, and thus improved kinetic behaviour. However, the excessively large specific surface area of the 4S-LiMn₂O₄-HMSs is a double-edged sword since it can cause the drastic dissolution of the surface Mn, inducing the Jahn–Teller effect and active material loss.^35,36 The Jahn–Teller effect will give rise to structural change, resulting in the increase of the Li-insertion/extraction impedance. As Fig. S9† shows, the R_ct of both the types of multi-shelled hollow microspheres increases after 100 cycles, whereas that of the 4S-HMSs increases more and even becomes larger than that of the 3S-HMSs. Together with the capacity loss caused by active material dissolution, the 4S-LiMn₂O₄-HMSs show a poorer cycling stability. The 3S-LiMn₂O₄-HMSs, which have a moderate specific surface area with the thick outer shell protecting the inner shells from direct electrochemical dissolution, exhibit a much better cycling stability. After 100 cycles, the discharge specific capacities of triple- and quadruple-shelled are 97.9 and 92.1 mA h g⁻¹, respectively. Impressively, even after 200 consecutive cycles, the capacity of 3S-HMSs still remains 91.2 mA h g⁻¹ with a high coulombic efficiency of over 95% (Fig. S10†), demonstrating a fading rate of only 0.10% per cycle, which is superior to all the reported results based on the LiMn₂O₄ cathode without carbon-coating or ion-doping (Table S1†). To further testify the structural integrity of 3S-HMSs, samples after cycling were examined by SEM. It turns out that the morphology and structure of the 3S-LiMn₂O₄-HMSs remain good after 100 cycles (Fig. S11†).

Given that the rate capability is also very important for the practical application of LIBs, the cycling performances of the 3S-LiMn₂O₄-HMSs at different charge/discharge rates were studied (Fig. 3c). The specific capacity decreases along with the increase of the current rate, which is common among all the electrode materials. Even so, the 3S-LiMn₂O₄-HMSs can still deliver good capacity at a high current density of 10 C. It should also be noted that a high and steady capacity of over 103 mA h g⁻¹ can be attained when lowering the current rate back to 1 C, validating the indeed “breathable” structure of multi-shelled hollow microspheres for high performance LIB cathode materials.

Conclusions

In summary, uniform multi-shelled LiMn₂O₄ hollow microspheres with a high purity have been successfully synthesized through a hard-template method. By controlling the ratio of Li precursor and Mn precursor, the composition of the product is accurately controlled. Moreover, the structural parameters of the products such as the shell number, shell thickness and porosity are accurately controlled, by adjusting the heating rate and adsorption duration. When tested as cathode materials for LIBs, 3S-LiMn₂O₄-HMSs demonstrate a better cycling stability than all the reported results based on un-coated or un-doped LiMn₂O₄ (the fading rate is only 0.10% per cycle for 200 cycles). The superior lithium-storage is benefited from the merits of the multi-shelled hollow microstructures, including enhanced reaction activity, increased lithium-storage sites, shortened ion/electron diffusion path, and improved structural stability with multiple shells supporting each other and the exterior shell protecting the inner shells from electrochemical dissolution. Considering the facile synthesis and the improved performance, one can expect that these MS-LiMn₂O₄-HMSs will develop a new avenue for the next generation of LIBs with lower cost and better cycling stability.

Acknowledgements

This work was supported financially by the National Natural Science Foundation of China (no. 51172235, 21203201, 51202248, 21201167, 51272165, 51372245, 51302266, 51472244 and 21401199, 51362024) and the National Science Fund for Distinguished Young Scholars (no. 21325105).

References

1. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 1997, 276, 1395–1397.
2. P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J.-M. Tarascon, Nature, 2000, 407, 496–499.
3. D. Y. Kim, P. Muralidharan, H.-W. Lee, R. Ruffo, Y. Yang, C. K. Chan, H. L. Peng, R. A. Huggins and Y. Cui, Nano Lett., 2008, 8, 3948–3952.
4. M. M. Thackeray, P. J. Johnson, L. A. Depicciotto, P. G. Bruce and J. B. Goodenough, Mater. Res. Bull., 1984, 19, 179–184.
5. M. M. Thackeray and A. J. Dekock, J. Solid State Chem., 1988, 74, 414–418.
6. M. Jayalakshmi, M. Mohan Rao and F. Scholz, Langmuir, 2003, 19, 8403–8408.
7. J. Cabana, T. Valdés-Solís, M. R. Palacín, J. Oró-Solé, A. Fuertes, G. Marbàn and A. B. Fuertes, J. Power Sources, 2007, 166, 492–498.
8. J.-Y. Luo, Y.-G. Wang, H.-M. Xiong and Y.-Y. Xia, Chem. Mater., 2007, 19, 4791–4795.
9. F. Wu, N. Li, Y. Su, H. Shou, L. Bao, W. Yang, L. Zhang, R. An and S. Chen, Adv. Mater., 2013, 25, 3722–3726.
10. M.-J. Lee, S. Lee, P. Oh, Y. Kim and J. Cho, Nano Lett., 2014, 14, 993–999.
11. P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930–2946.
12. X. Zhao, M. V. Reddy, H. X. Liu, S. Ramakrishna, G. V. S. Rao and B. V. R. Chowdari, RSC Adv., 2012, 2, 7462–7469.
13. X. Y. Lai, J. E. Halpert and D. Wang, Energy Environ. Sci., 2012, 5, 5604–5618.
14. J. Qi, X. Y. Lai, J. Y. Wang, H. J. Tang, H. Ren, Y. Yang, Q. Jin, L. J. Zhang, R. B. Yu, G. H. Ma, Z. G. Su, H. J. Zhao and D. Wang, Chem. Soc. Rev., 2015, 44, 6749–6773.
15. X. W. Lou, L. A. Archer and Z. C. Yang, Adv. Mater., 2008, 20, 3987–4019.
16. M. Hu, Y. Tian, J. P. Wei, D. G. Wang and Z. Zhou, J. Power Sources, 2014, 247, 794–798.
17. Y. R. Zhong, M. Yang, X. L. Zhou and Z. Zhou, Mater. Horiz., 2015, 2, 553–566.
18. L. W. Su, Y. Jing and Z. Zhou, Nanoscale, 2011, 3, 3967–3983.
19. F. Zou, X. L. Hu, Z. Li, L. Qie, C. C. Hu, R. Zeng, Y. Jiang and Y. H. Huang, Adv. Mater., 2014, 26, 6622–6628.
20. X. N. Fu, Z. R. Chang, K. Chang, B. Li, H. W. Tang, E. Shangguan, X.-Z. Yuan and H. J. Wang, Electrochim. Acta, 2015, 178, 420–428.
21. X.-F. Yang, J.-H. Yang, K. Zaghib, M. L. Trudeau and J. Y. Ying, Nano Energy, 2015, 12, 305–313.
22. W. Xiang, E.-H. Wang, M.-Z. Chen, H.-H. Shen, S.-L. Chou, H. Chen, X.-D. Guo, B.-H. Zhong and X. L. Wang, Electrochim. Acta, 2015, 178, 353–360.
23. Z. X. Chen, L. F. Cao, L. Chen, H. H. Zhou, C. M. Zheng, K. Xie and Y. F. Kuang, J. Power Sources, 2015, 298, 355–362.
24. X. Y. Lai, J. Li, B. A. Korgel, Z. H. Dong, Z. M. Li, F. B. Su, J. Du and D. Wang, Angew. Chem., Int. Ed., 2011, 50, 2738–2741.
25. Z. H. Dong, X. Y. Lai, J. E. Halphert, N. L. Yang, L. X. Yi, J. Zhai, D. Wang, Z. Y. Tang and L. Jiang, Adv. Mater., 2012, 24, 1046–1049.
26. J. Y. Wang, N. L. Yang, H. J. Tang, Z. H. Dong, Q. Jin, M. Yang, D. Kisailus, H. J. Zhao, Z. Y. Tang and D. Wang, Angew. Chem., Int. Ed., 2013, 52, 6417–6420.
27. S. M. Xu, C. M. Hessel, H. Ren, R. B. Yu, Q. Jin, M. Yang, H. J. Zhao and D. Wang, Energy Environ. Sci., 2014, 7, 632–637.
28. H. Ren, R. B. Yu, J. Y. Wang, Q. Jin, M. Yang, D. Mao, D. Kisailus, H. J. Zhao and D. Wang, Nano Lett., 2014, 14, 6679–6684.
29. J. Y. Wang, H. J. Tang, H. Ren, R. B. Yu, J. Qi, D. Mao, H. J. Zhao and D. Wang, Adv. Sci., 2014, 1, 1400011.
30. J. Tarascon, W. McKinnon, F. Coowar, T. Bowmer, G. Amatucci and D. Guyomard, J. Electrochem. Soc., 1994, 141, 1421–1431.
31. W. Sun, F. Cao, Y. Liu, X. Zhao, X. Liu and J. Yuan, J. Mater. Chem., 2012, 22, 20952–20957.
32. S. Karaal, H. Kose, A. O. Aydin and H. Akbulut, Mater. Sci. Semicond. Process., 2015, 38, 397–403.
33. X. He, J. Wang, H. P. Jia, R. Kloepsch, H. D. Liu, K. Beltrop and J. Li, J. Power Sources, 2015, 293, 306–311.
34. Y.-M. Lin, P. R. Abel, A. Heller and C. B. Mullins, J. Phys. Chem. Lett., 2011, 2, 2885–2891.
35. I. M. Hung, Y.-C. Yang, H.-J. Su and J. Zhang, Ceram. Int., 2015, 41, S779–S786.
36. Y. H. Wang, L. Chen, Y. G. Wang and Y. Y. Xia, Electrochim. Acta, 2015, 173, 178–183.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5qi00213c

‡ These authors contributed equally.

---