Carbon nanofiber-templated mesoporous TiO₂ nanotubes as a high-capacity anode material for lithium-ion batteries

Abstract

The present communication demonstrates the synthesis of uniform mesoporous TiO₂ nanotubes via hydrolysis and condensation of tetrabutyl titanate (TBOT) using carbon nanofibers as templates followed by calcination under air. We further demonstrate the use of such TiO₂ nanotubes as an anode material for rechargeable lithium-ion batteries (LIBs) with high discharge capacity, good rate capability and excellent cycling performance. They deliver a high discharge capacity of 209.6 mA h g⁻¹ at a current density of 0.2 C in the voltage range 1.0–3.0 V. They can still deliver a discharge capacity of 108.1 and 88.8 mA h g⁻¹ with 100 and 93.2% capacity retention after 500 charge–discharge cycles under current densities of 2 and 4 C, respectively.

---

Rechargeable lithium-ion batteries (LIBs) have been the forerunner in portable and mobile applications due to their lightweight, relatively high specific energy and specific power.¹ Compared to bulk or micro scale materials, nanostructured ones provide shorter diffusion time and better battery performances.² Among these nanostructures, one-dimensional (1D) nanofibers benefit from a higher flux of lithium ion across the electrode/electrolyte interface leading to facile diffusion of cations.³ Anatase TiO₂ has emerged as a promising LIBs anode material due to its high theoretical capacity (167.5 mA h g⁻¹), flat operating potential, and low volume expansion during lithium intercalation/deintercalation leading to long cycle life and durability.⁴ Recent researches have shown that the use of hollow structure can further improve its electrochemical properties.^2,5 Herein, we demonstrate the first fabrication of uniform mesoporous TiO₂ nanotubes by hydrolysis and condensation of TBOT using electrospun carbon nanofibers as templates followed by calcination under air. As an anode material for LIBs, the TiO₂ nanotubes deliver a high discharge capacity of 209.6 mA h g⁻¹ at a current density of 0.2 C. They can still deliver a discharge capacity of 108.1 and 88.8 mA h g⁻¹ with 100 and 93.2% capacity retention after 500 charge–discharge cycles under current densities of 2 and 4 C, respectively.

Scanning electron microscopy (SEM) image of the carbon nanofibers (Fig. 1A) shows that they have a diameter of ∼280 nm and length of ∼2.0 μm. TiO₂ coating leads to an increase of the diameter up to ∼435 nm, as shown in Fig. 1B. Subsequent calcination in air gives uniform TiO₂ nanotubes (Fig. 1C). A further transmission electron microscopy (TEM) observation of one nanotube (Fig. 1D) reveals its porous nature. And the porous structures were derived from the voids between aggregated nanoparticles. The high-resolution TEM (HR-TEM) image taken from one nanoparticle (Fig. 1D, inset) reveals clear lattice fringes with interplanar spacing of 0.35 nm corresponding to the (101) plane of the anatase phase of TiO₂.⁶ The X-ray diffraction (XRD) pattern (Fig. 1E) clearly shows the peaks of anatase TiO₂ (JCPDS no. 21-1272).⁶ The corresponding selected area electron diffraction (SAED) pattern (Fig. S1†) indicates that the nanoparticles are polycrystalline anatase structure, in good agreement with the XRD results.⁷ The Brunauer–Emmett–Teller (BET) surface area and pore volume (Fig. 1F) of the TiO₂ nanotubes are calculated to be 137 m² g⁻¹ and 0.26 cm³ g⁻¹, respectively. The pore size distribution analysis indicates that the TiO₂ nanotubes have average pore of 7.2 nm. The BET surface of Degussa P-25 (with minor rutile fraction) was also calculated to be 89.4 m² g⁻¹ (Fig. S2†), which is smaller than that of the TiO₂ nanotubes. All these observation indicate the formation of mesoporous TiO₂ nanotubes.

Fig. 1 SEM images of (A) carbon nanofibers, (B) TiO₂-coated carbon nanofibers, and (C) TiO₂ nanotubes. (D) TEM image of one TiO₂ nanotube (inset: HR-TEM image). (E) XRD pattern and (F) N₂ adsorption–desorption isotherm and pore size distribution (inset) of TiO₂ nanotubes.

Electrochemical measurements were carried out to evaluate the lithium-storage capabilities of mesoporous TiO₂ nanotubes. Fig. 2A shows the cyclic voltammograms (CVs) of TiO₂ nanotubes at a scan rate of 0.5 mV s⁻¹ between 1 and 3 V. In the first cycle, a pair of current peaks at 1.60 and 2.19 V are observed during the cathodic and anodic sweeps, respectively, which correspond to the insertion/deinsertion processes of Li⁺ ions in the TiO₂ framework. The cathodic peak shifts to a higher voltage of 1.63 V in the following cycles, whereas the anodic peak shifts to 2.18 V. The gradually increased intensity of the anodic peak also indicates less trapping of lithium in subsequent cycles. The sharp constructive and overlapping peak potential indicates excellent reversibility during electrochemical cycling and corresponds to a two-phase reaction mechanism: TiO₂ + xLi⁺ + xe⁻ ↔ LixTiO₂. Fig. 2B shows the galvanostatic charge–discharge voltage profiles of TiO₂ nanotubes electrode at a current density of 0.2 C (1 C = 167.5 mA h g⁻¹). The initial discharge capacity of the TiO₂ nanotubes electrode can reach 209.6 mA h g⁻¹, with a corresponding charge capacity of 176.6 mA h g⁻¹. It can be clearly seen that the initial charge curve is the reverse process of the discharge curve and the reversible efficiency is as high as 84%. In the second cycle, the insertion and extraction capacities are observed to be 181.3 and 175.3 mA h g⁻¹. The irreversible capacity of 6 mA h g⁻¹ is much smaller than that (ca. 33 mA h g⁻¹) of the initial cycle. This value is nearly maintained in the following cycles, showing that the irreversible loss is diminishing rapidly upon cycling. Previous studies observed that the anatase TiO₂ electrode materials have a first irreversible capacity loss about 10 to 40%.^5c,8,9 In order to confirm the rate performance, the galvanostatic charge–discharge experiments were further performed at different current rates from 0.2 to 2 C of different TiO₂ nanotube electrodes (Fig. 2C). The reversible charge capacities of 161.3, 138.4 and 113.5 mA h g⁻¹ (96.3, 82.6 and 67.8% of the theoretical capacity) were obtained at the rates of 0.5, 1 and 2 C, respectively. This high rate performance is comparable to that of the reported TiO₂ electrodes.^2,5b,d,f Based on the charge–discharge curves, we claim that our TiO₂ nanotubes are promising for application in high power LIB anodes.

Fig. 2 (A) CVs of mesoporous TiO₂ nanotubes electrode at a scan rate of 0.5 mV s⁻¹, (B) galvanostatic charge–discharge voltage profiles at current density of 0.2 C, and (C) galvanostatic charge–discharge curves at various current densities (0.2 to 2 C) of different TiO₂ nanotubes electrodes.

The anatase-dominant Degussa P-25 electrode was fabricated for the comparison purposes. Fig. 3A shows the SEM image of the commercial P-25 nanoparticles with the diameter of ∼40 nm. The cycling performances of TiO₂ nanotubes and P-25 nanoparticles electrodes are shown in Fig. 3B at different constant current rates, with Coulombic efficiency. Remarkably, the TiO₂ nanotubes electrode demonstrates excellent capacity retention at different rates. Specifically, reversible capacities of 108.1 and 88.8 mA h g⁻¹ with 100 and 93.2% capacity retention were obtained after 500 cycles at current rates of 2 and 4 C, respectively. The Coulombic efficiency is increased to over 99% within first several cycles and thereafter the same efficiency was maintained nearly 100% for 500 cycles, which indicates the excellent reversibility during cycling. After the deep cycling, we have carried out the SEM and TEM examination. Remarkably, the original textural properties of TiO₂ nanotubes can be well retained in terms of shape and structural integrity, indicating the good structural stability of this material (Fig. S3†). The charge capacity of P-25 electrode is 18.2 mA h g⁻¹ at the current density of 0.2 C, a 60% capacity loss during the 100 cycles. The rate capabilities of the TiO₂ nanotubes and P-25 nanoparticles electrodes were evaluated at various current rates (Fig. 3C). The initial discharge and charge capacities of the TiO₂ nanotubes electrode were about 209.1 and 177 mA h g⁻¹, respectively. As expected, the capacity decrease gradually as the current rate increases. At a high current rate of 10 C, TiO₂ nanotubes electrode still deliver a reversible capacity of about 42.8 mA h g⁻¹. More importantly, the capacity of our sample return to the original values when the rate is reduced back to 0.2 C. At all rate conditions, the TiO₂ nanotubes electrode showed higher charge–discharge capacities and much better capacity retention than the P25 electrode. In order to find the improvement factor for the high rate TiO₂ nanotubes electrode, electrochemical impedance spectroscopy (EIS) analysis was conducted (Fig. 3D). It is found that the TiO₂ nanotubes electrode shows a much smaller radius of semicircle in the Nyquist plots as compared to that of the P-25 electrode, suggesting the TiO₂ nanotubes exhibit higher conductivity. This indicates that the Li⁺ intercalation/deintercalation rate for the TiO₂ nanotubes is much higher than that of the P-25. The higher capacity, better rate capabilities and excellent cyclability of TiO₂ nanotubes reported here are comparable to the previous reported TiO₂ electrodes.^2,5b,d,f

Fig. 3 (A) TEM image of the Degussa P-25 nanoparticles, (B) cycling performance of TiO₂ nanotubes and P-25 nanoparticles electrodes and Coulombic efficiency of TiO₂ nanotubes electrode at a current rate of 2 C, (C) galvanostatic charge–discharge capacities of TiO₂ nanotubes and P-25 nanoparticles electrodes at various current densities, and (D) Nyquist plots from electrochemical impedance spectroscopy for TiO₂ nanotubes and P-25 electrodes.

The excellent electrochemical performances of the mesoporous TiO₂ nanotubes could be attributed to the following three reasons: (1) uniform TiO₂ nanotubes can be mixed homogeneously with conductive carbon, leading to good electronic conduction; (2) the high surface area of TiO₂ nanotubes constructs a high electrode/electrolyte interface, where the charge-transfer process occurs; (3) the mesopores formed between the TiO₂ nanoparticles facilitate the electrolyte transport.

In summary, we have developed uniform mesoporous TiO₂ nanotubes via hydrolysis and condensation of TBOT using electrospun carbon nanofibers as templates followed by calcination under air. As an anode material for LIBs, such nanotubes were found to exhibit high discharge capacity, good rate capability and excellent cycling performance. Thus, the mesoporous nanotubes hold great promise for application in high power LIB anodes.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21175129), the National Basic Research Program of China (no. 2011CB935800), and the Scientific and Technological Development Plan Project of Jilin Province (no. 20100534).

Notes and references

1. (a) M. Gratzel, Nature, 2001, 414, 338; (b) P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930.
2. X. Zhang, V. Aravindan, P. Suresh Kumar, H. Liu, J. Sundaramurthy, S. Ramakrishna and S. Madhavi, Nanoscale, 2013, 5, 5973.
3. S. Cavaliere, S. Subianto, I. Savych, D. J. Jones and J. Roziere, Energy Environ. Sci., 2011, 4, 4761.
4. (a) Z. Yang, D. Choi, S. Kerisit, K. Rosso, D. Wang, J. Zhang, G. Graff and J. Liu, J. Power Sources, 2009, 192, 588; (b) L. Kavan, Chem. Rec., 2012, 12, 131.
5. (a) X. Lou and L. A. Archer, Adv. Mater., 2008, 20, 1853; (b) J. Chen, L. Yan and Y. Tang, J. Power Sources, 2010, 195, 6893; (c) B. Song, S. Liu, J. Jian, M. Lei, X. Wang, H. Li, J. Yu and X. Chen, J. Power Sources, 2008, 180, 869; (d) H. Han, T. Song, J. Bae, L. Nazar, H. Kim and U. Paik, Energy Environ. Sci., 2011, 4, 4532; (e) Y. Wang, X. Su and S. Lu, J. Mater. Chem., 2012, 22, 1969; (f) H. Park, T. Song, H. Han, A. Devadoss, J. Yuh, C. Choi and U. Paik, Electrochem. Commun., 2012, 22, 81.
6. H. Wu, J. Chen, X. Lou and H. Hng, Nanoscale, 2011, 3, 4082.
7. S. Dai, Y. Wu, T. Sakai, Z. Du, H. Sakai and M. Abe, Nanoscale Res. Lett., 2010, 5, 1829.
8. J. Chen, H. Liu, S. Qiao and X. Lou, J. Mater. Chem., 2011, 21, 5687.
9. G. Kim, C. Jo, W. Kim, J. Chun, S. Yoon, J. Lee and W. Choi, Energy Environ. Sci., 2013, 6, 2932.

---

Footnote

† Electronic supplementary information (SI) available: Experimental section. See DOI: 10.1039/c3ra47239f

---