In situ synthesis of high-loading Li₄Ti₅O₁₂–graphene hybrid nanostructures for high rate lithium ion batteries

Abstract

Nanocrystalline Li₄Ti₅O₁₂ grown on conducting graphene nanosheets (GNS) with good crystallinity was investigated as an advanced lithium-ion battery anode material for potential large-scale applications. This hybrid anode nanostructure material showed ultrahigh rate capability and good cycling properties at high rates.

---

Recently, there has been a great interest in developing rechargeable lithium ion batteries (LIBs) for applications in automobiles and stationary power storage.^1,2 Although LIBs have gained commercial success and conquered the portable market, their implementation into electric transportation keeps being postponed due to low power, high cost and safety issues.^3,4 The key to overcoming these problems, essential if lithium-ion batteries are to be used for large-scale applications, lies in novel electrode materials with better electrochemical performance.⁵

The spinel Li₄Ti₅O₁₂ has become a highly promising anode material for LIBs. Lithium titanate-based materials show a higher lithium intercalation-deintercalation potential, which can effectively prevent safety problems associated with carbon-based anodes due to metallic formation during over-charging processes.^6,7 Furthermore, the spinel Li₄Ti₅O₁₂ possesses excellent reversibility, structural stability and excellent lithium ion mobility in the charge-discharge process.^8–10 However, Li₄Ti₅O₁₂ has inherently low electronic conductivity, which seriously limits its high rate capability.^11,12 Various approaches have been used to enhance its electron conductivity, such as surface coatings with conductive material^11–13 and doping aliovalent metal ions.^14,15 Recently, nanostructured Li₄Ti₅O₁₂ has appeared to be interesting because its large surface area increases the electrolyte/electrode contact area, which leads to a decrease of the current density per unit surface area and an increase in the charge/discharge rate.¹⁶ Also, the nanostructure provides a shorter path for lithium ion and electron transport and results in improved kinetic performance. Therefore the hybrid nanostructure electrodes composed of conductive additive nanophases and controlled particle growth is an ideal material both for rapid electronic and ionic transport, which are required for good rate capability.

However, the fabrication of nanosized Li₄Ti₅O₁₂ is still challenging when using the conventional synthesis routes because it requires further post-heat-treatment processing at high temperatures to achieve a high degree of crystallinity. Such tedious processes not only increase the manufacturing cost but also lead to unwanted particle growth. A good method to prepare nanocrystalline Li₄Ti₅O₁₂ is a chemical wet synthesis technique. Our group have used this method to synthesize hierarchically porous Li₄Ti₅O₁₂ microspheres assembled by very crystalline nanoparticles, which show enhanced charge transfer kinetics.¹⁷Graphene, a new two-dimensional carbon material, was recently considered as a potential alternative for graphite in LIBs because of its superior electrical conductivity, high surface area of over 2600 m² g⁻¹, chemical tolerance, ultrathin thickness and structural flexibility.^18–24

In this communication, we report a practical and efficient strategy for synthesizing Li₄Ti₅O₁₂/GNS nanocomposites, where crystalline Li₄Ti₅O₁₂ nanoparticles (NPs) with an optimal particle size are homogeneously anchored on conducting GNS. The overall fabrication procedures of Li₄Ti₅O₁₂/GNS nanocomposites are schematically illustrated in Scheme 1. The TiO₂ NPs were first anchored on graphite oxide (GO) by controlled hydrolysis of tetrabutyl titanate in the presence of GO. Then, the TiO₂/GO was in situ transformed into the Li₄Ti₅O₁₂ precursor (L-T-O) and a highly conductive form of graphene by hydrothermal process in the presence of hydrazine hydrate. Finally, the L-T-O/GNS was converted to Li₄Ti₅O₁₂/GNS nanocomposites by a short post-annealing. The experimental details are presented in the ESI.†The basic point in our paper is the mutually beneficial, i.e., symbiotic, role of the two intimately connected phases, GNS and Li₄Ti₅O₁₂. In these hybrid nanostructured electrodes, the ultrathin flexible GNS creates a two-dimensional conducting network for quickly providing electrons to Li₄Ti₅O₁₂ to assist in its storage and hinders the agglomeration and growth of Li₄Ti₅O₁₂ NPs during the calcination process, which reduces the transport path lengths of lithium ions and electrons. In turn, the Li₄Ti₅O₁₂ NPs attached on GNS can effectively prevent the aggregation and restacking of GNS to keep their large active contact area between the electrode and electrolyte. Therefore, the Li₄Ti₅O₁₂/GNS hybrid nanostructures can provide fast ionic and electronic conduction, resulting in high rate performance.

Scheme 1 Schematic illustration of the synthesis and the structure of Li₄Ti₅O₁₂/GNS nanocomposites

Fig. S1 of the ESI†shows the typical X-ray diffraction (XRD) patterns of the as-prepared Li₄Ti₅O₁₂ and Li₄Ti₅O₁₂/GNS composite. In Fig. S1a of the ESI,†all the diffraction peaks are ascribed to the pure phase of well-crystallised Li₄Ti₅O₁₂. As displayed in Fig. S2 of the ESI,†one intense peak at 10.6° is observed, which corresponds to the (002) diffraction peak of GO, and the interlayer spacing (0.87 nm) was much larger than that of pristine graphite (0.34 nm) due to the introduction of oxygen-containing functional groups on the graphite sheets. In Fig. S1b of the ESI,†the typical diffraction peak of GO disappears, whereas a small and low broad (002) diffraction peak appears at a 2θ value of 26°, which indicates that the layer-stacking regularity almost disappeared after reduction by hydrazine hydrate.²² All of the other diffraction peaks can be ascribed to the well-crystallised Li₄Ti₅O₁₂. That GO has been successfully reduced to graphene is further confirmed by Raman spectroscopy. The Raman spectrum of Li₄Ti₅O₁₂/GNS is shown in Fig. S3†, broad peaks displayed at 1331 cm⁻¹ and 1592 cm⁻¹ are assigned to the D and G peaks of graphene. A stronger D band was observed than G band, indicating a largely disordered structure of the obtained GNS.¹⁹ The result obtained by TG analysis indicates that the amount of GNS in the Li₄Ti₅O₁₂/GNS nanocomposites was about 15.5 wt.% (Fig. S4 of the ESI†).

The GNS are entangled with each other and resemble crumpled paper, as shown in Fig. S5a of the ESI.†Corrugation and scrolling are part of the intrinsic nature of GNS, which result from the fact that the 2D membrane structure becomes thermodynamically stable via bending.^19,20 Fig. S5b of the ESI† is a (TEM) image of the TiO₂/GO, which shows the presence of TiO₂ NPs with an average diameter of 10 nm distributed on the exfoliated GO nanosheets. The GO sheets decorated with distributed functional groups can act as anchor sites and consequently make the in situ formed NPs attach to the surface and edges of the GO sheets.²⁵ The Li₄Ti₅O₁₂/GNS was obtained from the hydrothermal treatment of TiO₂/GO in LiOH solution containing a small amount of hydrazine hydrate and following the calcination procedure. This process can ensure the in situ formation of Li₄Ti₅O₁₂ NPs and GNS simultaneously, which can ensure that the Li₄Ti₅O₁₂ NPs are homogeneously anchored on graphene layers and prevent any serious stacking of GNS, avoiding/weakening the loss of their high active surface area. It can be seen from Fig. 1 (a, b) that the Li₄Ti₅O₁₂ NPs are uniformity attached on the graphene layers, and the GNS were separated by the Li₄Ti₅O₁₂ NPs. It is worth noting that during the hydrothermal process, the NPs are still strongly anchored on the surface of GNS with a high density, suggesting a strong interaction between NPs and GNS. The strong anchoring of Li₄Ti₅O₁₂ NPs on GNS enables rapid electron transport through the underlying GNS to NPs, resulting in superior rate capability. Fig. 1 (c) shows a TEM image of the as-prepared Li₄Ti₅O₁₂/GNS nanocomposites, and a very large quantity of Li₄Ti₅O₁₂ NPs with a size of 30 nm are uniformly dispersed onto/into the GNS. A lattice resolved high resolution transmission electron microscope (HRTEM) image of the Li₄Ti₅O₁₂ NPs and GNS is presented in Fig. 1 (d). The stacking of GNS is of about 2–6 layers with a (002) interplanar distance of 0.38 nm, which is significantly larger than that in pristine graphite (0.34 nm). A lattice fringe of Li₄Ti₅O₁₂ about 0.48 nm in size is clearly observed, which is in good agreement with the spacing of the (111) planes of spinel Li₄Ti₅O₁₂, thus demonstrating the highly crystalline nature of the Li₄Ti₅O₁₂ nanocrystals. Atomic force microscopy (AFM) analysis was further conducted to investigate the structural features of Li₄Ti₅O₁₂/GNS nanocomposites (Fig. S6†). The thickness of the GNS is about 2 nm. The peak marked in the profile corresponds to the height of Li₄Ti₅O₁₂ nanoparticle on GNS (about 30 nm). The results of the AFM observations are consistent with the TEM results. It should be emphasised that the product prepared in the absence of GNS consisted of only irregular NPs with a size of 100 nm (Fig. S7 of the ESI†), which is much larger than that of Li₄Ti₅O₁₂ NPs in the Li₄Ti₅O₁₂/GNS composite. This result strongly indicates that the graphene is the key factor in controlling the morphology by acting as the nucleation site for deposits through interaction with metal ions.

Fig. 1 (a, b) Field emission scanning electron microscope (FESEM), (c) TEM and (d) HRTEM images of Li₄Ti₅O₁₂/GNS nanocomposites.

To examine the effectiveness of GNS in improving the rate capability of the electrode, the Li-ion insertion/extraction properties in the Li₄Ti₅O₁₂/GNS composite and pure Li₄Ti₅O₁₂ were investigated, respectively. The charge-discharge curves of the Li₄Ti₅O₁₂ electrode at different current rates from 0.1 to 10 C are shown in Fig. 2 (a). The first discharge capacity is 168.5 mAh g⁻¹ at a rate of 0.1 C. At the rate of 10 C, the pure Li₄Ti₅O₁₂ has no clear discharge voltage plateau, and the capacity is only 81.2 mAh g⁻¹. The greatly improved electrochemical performance of the Li₄Ti₅O₁₂/GNS electrode is shown in Fig. 2 (b). The Li₄Ti₅O₁₂/GNS electrode delivers a high specific capacity of 171.4 mAh g⁻¹ at 0.1 C (the capacity is calculated based on the weight of the Li₄Ti₅O₁₂ in the Li₄Ti₅O₁₂/GNS electrode). Although the cell voltage decreases with increasing current density, it also shows a flatter potential plateau even at 30 C. Fig. 2 (c) compares the rate capabilities of Li₄Ti₅O₁₂/GNS and Li₄Ti₅O₁₂ electrodes at different rates. The discharge capacities of Li₄Ti₅O₁₂ decrease steeply with increasing discharge rate, whereas Li₄Ti₅O₁₂/GNS nanocomposites decrease much slower at the same rate. It is noteworthy that the capacity (82.7 mAh g⁻¹) obtained by Li₄Ti₅O₁₂/GNS nanocomposites at a rate of 60 C is higher than that obtained at a rate of 10 C for the pure Li₄Ti₅O₁₂. A significantly improved rate capability was clearly achieved for the Li₄Ti₅O₁₂/GNS nanocomposites. The Li₄Ti₅O₁₂/GNS nanocomposites also exhibited excellent cyclability with no noticeable decrease in performance over 100 cycles (Fig. 2 (d)). The discharge capacity loss was 3.1% at 1 C, 2.5% at 10 C and 2.9% at 60 C. This result demonstrates that the structure of the composite is very stable, and the electrochemical Li⁺ insertion/extraction process is quite reversible even at high rates. However, for the pure Li₄Ti₅O₁₂, the discharge capacity loss was more than 4% at 1 C and even 7% at 10 C (Fig. S8 of the ESI†). It is reasonable to assume that the excellent cycling stability of the Li₄Ti₅O₁₂/GNS nanocomposites is related to their good crystallinity and the improvement of the electrical conductivity.

Fig. 2 (a) Capacity–voltage profile for Li₄Ti₅O₁₂. (b) Capacity–voltage profile for Li₄Ti₅O₁₂/GNS nanocomposites. (c) Comparison of the rate capabilities of Li₄Ti₅O₁₂/GNS nanocomposites and pure Li₄Ti₅O₁₂ at different rates. (d) Cycle performance of the Li₄Ti₅O₁₂/GNS electrode for different current densities.

To understand the improved high-rate performance after introducing conducting GNS into Li₄Ti₅O₁₂, we obtained electronic conductivity and electrochemical impedance spectroscopy (EIS) measurements for the Li₄Ti₅O₁₂/GNS hybrid materials and the pure Li₄Ti₅O₁₂. The electronic conductivity of the Li₄Ti₅O₁₂/GNS hybrid materials (ca. 0.0041 S m⁻¹) is about a factor of 10⁸ higher than the pure Li₄Ti₅O₁₂ (ca. 10⁻¹¹ S m⁻¹), demonstrating the improved electron transport due to the existence of the superior electrical conductivity of graphene. The EIS result (Fig. S9 of the ESI†) shows that each curve consists of a depressed semicircle in the high-middle frequency region and an oblique straight line in the low frequency region. In Table S1 of the ESI,† the Li₄Ti₅O₁₂/GNS nanocomposite electrodes exhibited much lower charge-transfer resistance than that of the Li₄Ti₅O₁₂ electrode. Furthermore, the exchange current densities (i⁰ = RT/nFRct)²⁶ of the Li₄Ti₅O₁₂/GNS cell were higher than those of the Li₄Ti₅O₁₂ cell. Therefore, the improved high rate performance may be attributed to three aspects: (1) The GNS work as a highly conductive matrix for quickly providing electrons to Li₄Ti₅O₁₂ to assist its storage. (2) The GNS can provide a support for anchoring well-dispersed Li₄Ti₅O₁₂, which hinders Li₄Ti₅O₁₂ NPs agglomeration and growth during the calcination process and thus reduces the transport path lengths of lithium ions and electrons. (3) The Li₄Ti₅O₁₂ NPs are well dispersed onto GNS as spacers to effectively reduce the degree of restacking of GNS and consequently keep their large active contact area between the electrode and electrolyte.

In summary, we developed a facile route to fabricate a composite of electrically conductive GNS anchored with nanocrystalline Li₄Ti₅O₁₂ as an advanced anode material for high performance LIBs. The hybrid nanostructures endow the composites with high-rate transportation for both Li⁺ and electrons (especially at high rates) because the hybrid nanostructure is capable of effectively utilising the good conductivity, high surface area, good electrochemical performance of GNS and the good stability of fine Li₄Ti₅O₁₂ NPs. It is the synergy of the two parts that leads to a high rate capability, high rate and stable lithium storage material.

Acknowledgements

This work was supported by the National Basic Research Program of China (973 Program) (No. 2007CB209703), and the National Natural Science Foundation of China (No.206033040, No. 20873064).

References

1. B. Scrosati, Nature, 1995, 373, 557–558.
2. A. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon and W. V. Schalkwijk, Nat. Mater., 2005, 4, 366–377.
3. J. M. Tarascon and M. Armand, Nature, 2008, 451, 652–657.
4. P. Gibot, M. C. Cabanas, L. Laffont, S. Levasseur, P. Carlach, S. Hamelet, J. M. Tarascon and C. Masquelier, Nat. Mater., 2008, 7, 741–747.
5. G. Armstrong, A. R. Armstrong, P. G. Bruce, P. Reale and B. Scrosati, Adv. Mater., 2006, 18, 2597–2600.
6. K. S. Park, A. Benayad, D. J. Kang and S. G. Doo, J. Am. Chem. Soc., 2008, 130, 14930–14931.
7. Y. G. Wang, H. M. Liu, K. X. Wang, H. Eiji, Y. R. Wang and H. S. Zhou, J. Mater. Chem., 2009, 19, 6789–6795.
8. T. Ohzuku, A. Ueda and N. Yamamoto, J. Electrochem. Soc., 1995, 142, 1431–1435.
9. E. M. Sorensen, S. J. Barry, H. K. Jung, J. R. Rondinelli, J. T. Vaughey and K. R. Poeppelmeier, Chem. Mater., 2006, 18, 482–489.
10. L. F. Shen, C. Z. Yuan, H. J. Luo, X. G. Zhang, K. Xu and F. Zhang, J. Mater. Chem., 2011 10.1039/C0JM02316G.
11. L. Cheng, J. Yan, G. N. Zhu, J. Y. Luo, C. X. Wang and Y. Y. Xia, J. Mater. Chem., 2010, 20, 595–601.
12. G. J. Wang, J. Gao, L. J. Fu, N. H. Zhao, Y. P. Wu and T. Takamura, J. Power Sources, 2007, 174, 1109–1112.
13. A. S. Prakash, P. Manikandan, K. Ramesha, M. Sathiya, J. M. Tarascon and A. K. Shukla, Chem. Mater., 2010, 22, 2857–2863.
14. D. Capsoni, M. Bini, V. Massarotti, P. Mustarelli, S. Ferrari, G. G. Chiodelli, M. C. Mozzati and P. Galinetto, J. Phys. Chem. C, 2009, 113, 19664–19671.
15. J. Wolfenstine and J. L. Allen, J. Power Sources, 2008, 180, 582–585.
16. D. Deng, M. G. Kim, J. Y. Lee and J. Cho, Energy Environ. Sci., 2009, 2, 818–837.
17. L. F. Shen, C. Z. Yuan, H. J. Luo, X. G. Zhang, K. Xu and Y. Y. Xia, J. Mater. Chem., 2010, 20, 6998–7004.
18. E. Yoo, J. Kim, E. Hosono, H. Zhou, T. Kudo and I. Honma, Nano Lett., 2008, 8, 2277–2282.
19. G. X. Wang, X. P. Shen, J. Yao and J. Park, Carbon, 2009, 47, 2049–2053.
20. D. Y. Pan, S. Wang, B. Zhao, M. H. Wu, H. J. Zhang, Y. Wang and Z. Jiao, Chem. Mater., 2009, 21, 3136–3142.
21. J. Yao, X. P. Shen, B. Wang, H. K. Liu and G. X. Wang, Electrochem. Commun., 2009, 11, 1849–1852.
22. Z. S. Wu, W. C. Ren, L. Wen, L. B. Gao, J. P. Zhao, Z. P. Chen, G. G. Zhou, F. Li and H. M. Cheng, ACS Nano, 2010, 4, 3187.
23. D. H. Wang, D. W. Choi, J. Li, Z. G. Yang, Z. M. Nie, R. Kou, D. H. Hu, C. M. Wang, L. V. Saraf, J. G. Zhang, I. A. Aksay and J. Liu, ACS Nano, 2009, 3, 907–914.
24. N. Zhu, W. Liu, M. Q. Xue, Z. Xie, D. Zhao, M. N. Zhang, J. T. Chen and T. B. Cao, Electrochim. Acta, 2010, 55, 5813–5818.
25. G. X. Wang, B. Wang, X. L. Wang, J. Park, S. X. Dou, H. Ahn and K. Kim, J. Mater. Chem., 2009, 19, 8378–8384.
26. X. Li, M. Z. Qu, Y. J. Huai and Z. L. Yu, Electrochim. Acta, 2010, 55, 2978–2982.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, XRD, TG, TEM, SEM, EIS. See DOI: 10.1039/c0nr00639d

---