TiO2 nanotubes grown on graphene sheets as advanced anode materials for high rate lithium ion batteries

Yufeng Tanga, Zhanqiang Liua, Xujie Lüa, Baofeng Wangc and Fuqiang Huang*ab
aCAS Key Laboratory of Materials for Energy Conversion and State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China. E-mail: huangfq@mail.sic.ac.cn; Fax: +86 21 52416360; Tel: +86 21 52411620
bState Key Laboratory of Rare Earth Materials Chemistry and Applications, College of Chemistry & Molecular Engineering, Peking University, Beijing 100871, China
cCollege of Environmental and Chemical Engineering, Shanghai University of Electric Power, Shanghai 200090, China

Received 9th July 2014 , Accepted 31st July 2014

First published on 31st July 2014


Abstract

Here, we propose a unique architecture based on directly growing TiO2 nanotubes on graphene sheets. This work establishes the benefits of combining nanotubular TiO2 with conductive graphene to achieve a significant enhancement in capacitive energy storage. The TiO2 nanotubes possess a porous structure and a higher surface area compared with its bulk material. The graphene sheets play an important role in the conductive substrate for charge collection. In the preparation process, amorphous TiO2 colloids coated on graphene oxide (GO) sheets serve as the seeds to realize the growth of nanotubes, and ensure the robust contact between graphene and TiO2 nanotubes to benefit charge transfer. When used as anodes for lithium ion batteries, the graphene–TiO2 nanotubes (Gr–TNTs) manifest excellent rate capability and cycling performance, indicating that the nanotubes, graphene and their synergistic effect can prompt fast charge transfer in such a composite structure.


Introduction

Graphene possesses superior electrical conductivity, a large specific surface area, strong mechanical stability, and chemical stability.1,2 Designing and fabricating graphene-based nanocomposites with highly specific morphology and composition has become a hot topic for research in modern materials chemistry because of their new and/or enhanced functionalities, which cannot be achieved by either component alone, and their wide application in electronic, photocatalytic and energy storage fields.3,4 In this regard, remarkable progress has been made in hybrid nanocomposites consisting of graphene with functional materials, such as metals oxide,5–7 copolymers,8,9 and biomaterials.10

Titanium oxide is one of the most intensively investigated transition-metal oxides as a lithium-ion host because of its high theoretical capacity, high safety, environmental benignity, etc.11–13 A challenging issue in using TiO2 for lithium-ion batteries is to tackle its poor electronic conductivity. Early research work manifested that the introduction of a carbon-based conductor or nanoarchitectured electrodes is an effective strategy to improve electronic conduction. Owing to its superior conductivity and larger surface area than a conventional carbon conductor, a great effort has recently been focused on constructing graphene-based hybrid composites with nanostructured TiO2 (including nanoparticles, nanorods and nanosheets) for improving lithium storage performance.13–15 Although many kinds of graphene–TiO2 nanocomposites have been made, it is hard to build up a good interface between graphene and TiO2 nanotubes, which is required for rapid ion diffusion and electron transition during the Li+ insertion–extraction process.

Herein, a unique architecture of graphene-based hybrid TiO2 nanotubes was fabricated using GO sheets decorated with TiO2 colloids as a precursor. TiO2 colloids on GO, served as seeds to initialize the growth of TiO2 nanotubes and ensured the tight bonding between the graphene sheets and nanotubes. Because of the short ion diffusion path of nanotubes and the high conductivity of graphene, the Gr–TNTs nanocomposites should be an ideal anode material for high rate lithium ion batteries. Galvanostatic testing results demonstrate that the nano-architectured Gr-TNTs show enhanced Li+ insertion–extraction kinetics in TiO2, especially at high charge–discharge rates.

Experimental section

Preparation of Gr–TNTs nanocomposites

For the fabrication of Gr–TNTs, TiO2 colloids coated on graphene oxide sheets were first prepared by the controlled hydrolysis of titanium tetraisopropoxide (TTIP, 97% Aldrich) in the presence of GO sheets. Then the TiO2 colloids-contained GO sheets were in situ transformed into Gr–TNTs by a hydrothermal process followed by calcination. Typically, 40 mg of GO, fabricated using the modified Hummers method,16 was dispersed into 20 mL of an ethanol–water (50[thin space (1/6-em)]:[thin space (1/6-em)]1) solution, followed by the addition of 2 mmol of TTIP with stirring. After being stirred for 20 min, 2 mL of distilled water was added to ensure complete hydrolysis of TTIP. The TiO2 colloids-contained GO deposited at the bottom of the vessel, which was collected and then mixed with 20 mL of NaOH (0.1 M) and H2O2 (0.04 wt%) solution. The suspension was transferred into a 30 mL autoclave and then placed in an oven at 180 °C for 8 h. After the hydrothermal reaction, the products were neutralized using 0.1 M HCl and subsequently washed with deionized water to pH = 7, dried at 60 °C under vacuum for 3 h, and finally treated at 400 °C in H2/Ar for 4 h with a heating rate of 1 °C min−1 to obtain the Gr–TNTs. For comparison, TiO2 nanotubes were prepared by a similar process without the addition of graphene.

Characterization

The structure and morphology of the products were characterized by X-ray diffraction measurements (XRD, Rigaku, D/max-RB using Cu Ka radiation), transmission electron microscopy (TEM, JEOL JEM-2010 equipped with an energy dispersive X-ray detector) and field emitting scanning electron microscopy (FESEM, JEOL JSM-7401F), respectively. High resolution TEM was performed on a JEOL JEM-2100F with an acceleration voltage of 200 kV. The N2 sorption studies were carried out at 77 K using a Micromeritics ASAP 2000 instrument. Raman spectra of the samples were obtained with Raman spectroscopy with a laser excitation energy of 532 nm. Thermogravimetric analysis (TGA) was conducted on a TA instruments Q50 by heating from room temperature to 700 °C at 5 °C min−1 in a nitrogen atmosphere.

Electrochemical characterization

The charge and discharge capacities were measured with coin cells in which a lithium metal foil was used as the counter electrode. The electrolyte employed was a 1 M solution of LiPF6 in ethylene carbonate and dimethyl carbonate (EC + DMC) (1[thin space (1/6-em)]:[thin space (1/6-em)]1 in volume). The Gr–TNTs powder (80 wt%), acetylene black (10 wt%) and polyvinylidene fluoride binder (10 wt%) were homogeneously mixed in NMP solvent with magnetic stirring (For pure TiO2 nanotubes, the ratio is 65[thin space (1/6-em)]:[thin space (1/6-em)]25[thin space (1/6-em)]:[thin space (1/6-em)]10). After stirring for 3.5 h, the slurry was coated uniformly on aluminum foil. Finally, the electrode was dried under vacuum at 110 °C for 10 h. Cell assembly was carried out in an argon-filled glove box (German, M. Braun Co., [O2] < 1 ppm, [H2O] < 1 ppm). The coin cells were cycled under different current densities between cut off voltages of 3.0 and 1.0 V on a CT2001A cell test instrument (LAND Electronic Co.) at room temperature.

Result and discussion

In a previous report, we proposed a mild hydrothermal approach to prepare TiO2 nanotubes. In that process, TiO2 colloids are used as seeds to initialize the growth of sodium titanate nanotubes in a low concentration solution of NaOH and H2O2, and the sodium titanate nanotubes converted to TiO2 nanotubes after acid washing and heat treatment. Our investigation reveals that H2O2 plays an important role in the formation of titanate nanotubes, since only titanate nanosheets are formed without the addition of this agent.17 With the assistance of H2O2, the titanate nanosheets can scroll into nanotubes even in a low concentration solution of NaOH (0.1 M). In the present work, we further developed this method to grow TiO2 nanotubes on graphene sheets.

As illustrated by Fig. 1a, GO sheets with tens of micrometers are used as the support for growth of TiO2 nanotubes. Since the functional groups of GO provide nucleation centers to initialize the seeds of TiO2, GO sheets can be uniformly coated by a layer of TiO2 colloids from the hydrolysis of TTIP in ethanol (Fig. 1b). Theses TiO2 colloids on GO could guide the growth of titanate nanotubes. During the hydrothermal process, homogenous tiantate nanotubes grow on the GO sheets to form a 2D plate-like structure (Fig. 1c and d and S1 in the ESI). After acid washing and calcination in H2/Ar, the 2D plate-like GO–titanate nanotubes composites transform to Gr–TNTs without their structure being destroyed. As shown in Fig. 2a, a piece of graphene is completely covered by TiO2 nanotubes. Such 2D plate-like structure of Gr–TNTs demonstrates the successful growth of TiO2 nanotubes on the graphene sheets, since only aggregation of TiO2 nanotubes will be formed without the graphene support (Fig. S2 in the ESI). The detailed morphology of the 2D structure of Gr–TNTs (Fig. 2b) shows that a dense mass of TiO2 nanotubes is uniformly distributed on the graphene sheets. These TiO2 nanotubes are apart from each other and the aggregation of nanotubes is retarded. Fig. 2c and d show the TEM images of Gr–TNTs, which further confirm that the roots of TiO2 nanotubes are firmly fixed on the graphene support. We suppose that the tight contact between nanotubes and graphene is favorable to improve the lithium storage performance. It can also be seen in Fig. 2d that the length of TiO2 nanotubes is about 200 nm, and the inner diameter is ca. 20 nm.


image file: c4ra05027d-f1.tif
Fig. 1 SEM images of (a) GO sheets, (b) GO sheets coated by TiO2 colloids, (c and d) SEM and TEM images of GO sheets supported titanate nanotubes.

image file: c4ra05027d-f2.tif
Fig. 2 (a and b) Scanning electron microscope images of Gr–TNTs; inset in (b) high-magnification image showing TiO2 nanotubes on the graphene sheets; (c and d) transmission electron microscope images of the Gr–TNTs.

X-ray diffraction (XRD) is used to monitor the structure and composition of the samples. As displayed in Fig. 3a, the diffraction peak at 10.6° corresponds to the (002) diffraction peak of GO.18 After hydrothermal reaction and following heat treatment, the typical diffraction peak of GO disappears, whereas a small and low broad (002) diffraction peak appears at 26°, which indicates that GO has been successfully reduced to graphene.13 For Gr–TNTs, all the diffraction peaks are ascribed to the anatase phase of TiO2. There are no obvious diffraction peaks of graphene observed in the pattern of Gr–TNTs, suggesting that graphene sheets are homogeneously dispersed and coated with TiO2 nanotubes.18


image file: c4ra05027d-f3.tif
Fig. 3 X-ray diffraction patterns of graphene, graphene oxide and the Gr–TNTs.

The graphene in the Gr–TNTs can be determined by Raman analysis. Fig. 4 shows the typical Raman spectra of anatase TiO2 and the Gr–TNTs. The TiO2 spectra display three Raman-active fundamentals in the vibrational spectrum: Eg modes centred around 639 cm−1, and two B1g modes at 399 and 519 cm−1. Besides, the Raman-allowed phonon peaks of TiO2, two peaks at higher frequencies (1331 cm−1 and 1592 cm−1) observed in the spectra of the Gr–TNTs can be assigned to the typical D-band and G-band peaks of graphene.7 The D/G intensity ratio of the Gr–TNTs decreased when compared to that in the spectrum of GO–TiO2, which suggests an increase in the average size of the in-plane sp2 domain upon the reduction of GO to graphene.


image file: c4ra05027d-f4.tif
Fig. 4 Raman spectra of the Gr–TNTs, anatase TiO2 and GO–TiO2 before the hydrothermal process.

The oxygen content of GO before and after reduction was analyzed by XPS (Fig. S3 in the ESI). The oxygen content of GO was estimated to be 38%, while after heat-treatment at 400 °C in H2/Ar for 4 h the percentage of oxygen decreased to 5.6%. FT-IR (Fig. S4 in the ESI) shows the characteristic peaks of GO appear for O–H (3400 and 1625 cm−1), C[double bond, length as m-dash]O (1721 cm−1), epoxy C–O (1219 cm−1) and alkoxy C–O (1053 cm−1) groups.19 Although the peaks for the C[double bond, length as m-dash]O groups are reduced significantly after reduction, the residual peaks for the alkoxy C–O groups are still found with a small shift from 1053 cm−1 to 1110 cm−1. These observations suggest that C–O functional groups may link to the TiO2 nanotubes, indicating the possible presence of C–O–Ti bonds between TiO2 and graphene. Similar deduction has also been done in previous reports on magnetite/graphene nanosheet composites.20 In addition, the peaks of Gr–TNTs at 500–900 cm−1 can be indexed to the Ti–O–Ti stretching vibration modes in the TiO2.21 The TG analysis indicates that the content of TiO2 in the Gr–TNTs is about 84.5 wt% (Fig. S5 in the ESI).

The specific surface area and pore structure of the Gr–TNTs was analyzed by nitrogen adsorption and desorption (shown in Fig. 5). The isotherm is a typical type IV-like with type H3 hysteresis loop, indicating the presence of mesoporous material. The Barrett–Joyner–Halenda (BJH) pore-size distribution curve indicates that the average pore diameter is 18.5 nm, which is consistent with the TEM observations. In addition, the Brunauer–Emmett–Teller (BET) surface area is 105 m2 g−1. It is well known that the large specific surface area and mesoporous structure of nanotubes are favorable for rapid lithium-ion insertion–extraction in electrode materials. Furthermore, the graphene substrate can enhance the electronic conductivity of TiO2. When used in lithium ion batteries, the high ionic and electronic conductivity will delay the capacity loss associated with the concentration polarization to high current density. Thus, the Gr–TNTs might be an ideal system of choice for application in high rate lithium ion batteries.


image file: c4ra05027d-f5.tif
Fig. 5 Nitrogen adsorption–desorption BET isotherm and the pore size distribution of the Gr–TNTs.

Recently, extensive work has been made in exploration of various graphene–nano-TiO2 composites in order to enhance capacity and cycle performance. Beyond that, the rate performance of TiO2 also needs to be improved in the light of future battery applications in high-power devices. The voltage profiles of Gr–TNTs composite for charging to 3.0 V and discharging to 1.0 V at increasing rates from 0.5 C to 10 C (1 C = 337 mA h g−1) are shown in Fig. 6a. They demonstrate that the Gr–TNTs nanocomposite can sustain excellent charge and discharge rates. At the low rate of 0.5 C, the discharge process of the sample consists of three stages; the first stage is the quick voltage drop, the second stage is the distinct voltage plateau, and the third stage is a gradual decay in potential. The voltage plateau is related to the phase transition between the tetragonal (I41/amd) and orthorhombic (Li0.5TiO2, space group Pnm21) phases with Li insertion into anatase TiO2. The voltage plateau decreases with an increasing current density. It is 1.75 V at 0.5 C and drops to about 1.25 V at 10 C. Moreover, the specific capacity exhibits a tendency to decrease with increasing current density.


image file: c4ra05027d-f6.tif
Fig. 6 (a) The first charge–discharge curves and (b) rate performance of the Gr–TNTs composite between 1.0 V and 3.0 V at different current densities.

The rate performance of the Gr–TNTs at different current densities is illustrated in Fig. 6b. It can be seen that a specific capacity of 223 mA h g−1 is obtained at 0.5 C after 5 cycles, and the value decreases gradually to 180 mA h g−1 at 1 C after 10 cycles, and 145 mA h g−1 at 5 C after 10 cycles. Even at the high rate of 10 C, a capacity of 118 mA h g−1 can still be delivered, which is not only superior to the capacity of bare TiO2 nanotubes (80 mA h g−1 at 10 C) but also higher than that of previously reported graphene–TiO2 electrodes.13–15,22–25 The large capacities of Gr–TNTs certify the much enhanced lithium storage performance of TiO2 nanotubes by incorporation of graphene sheets. When the charge–discharge current density returns to 1 C, the capacity of Gr–TNTs can be restored to 178 mA h g−1, showing a high reversibility. The excellent rate performance of Gr–TNTs should be ascribed to its unique hybrid structure and the synergistic effect of the components. On the one hand, the nanotubes with interspacing have a higher electrode–electrolyte contact area than aggregated TiO2 nanotubes, which provides more channels for Li-ion insertion–extraction and electron transition. On the other hand, the graphene sheets play an important role as a highly conductive substrate for current collection, which enhances the electron conductivity of the whole electrode. Besides, the tight bonding between the nanotubes and graphene sheets, which facilitates fast electron-transport to the current collector and enhances the rate capability of TiO2.

We also investigated the cycle behavior of the Gr–TNTs at the high rate of 5 C, as shown in Fig. 7. In the first cycle, the capacity of the sample decreases significantly from 176 to 159 mA h g−1. The serious capacity loss can be ascribed to irreversible Li insertion sites in TiO2 and the adsorbed trace water in the electrode, respectively.26 After the initial capacity loss, the Gr–TNTs composite shows very high capacity retention upon cycling. The specific capacity stays above 140 mA h g−1 after 150 cycles and the average capacity loss from the 2nd to 150th cycles is ca. 0.12 mA h g−1 per cycle. The exceptional cycle stability should be ascribed to the graphene sheets in the hybrid structure, which can accommodate the strains of TiO2 during lithium-ion insertion–extraction.


image file: c4ra05027d-f7.tif
Fig. 7 The cycle performance of the Gr–TNTs at 5 C in the potential range of 3.0–1.0 V.

Conclusions

In summary, we have designed a graphene-based hybrid structure by directly growing TiO2 nanotubes on graphene sheets. The synthesis of the Gr–TNTs is based on the hydrothermal conversion of TiO2 colloids-contained GO to TiO2 nanotubes on graphene. By taking the advantages of nanotubes and highly conductive graphene, the unique hybrid structure is favorable to rapid ion diffusion and electron transition. The electrochemical testing demonstrates the Gr–TNTs composite as a promising anode material for lithium ion batteries with high specific capacity, excellent rate performance, and good cycle stability.

Acknowledgements

This work is supported by National 973 & 863 Program of China Grant no. 2009CB939900 & 2011AA050505, NSF of China Grant no. 21203234 & 11274328 & 51125006 & 91122034 & 51121064 & 51102263 & 21101164, NSF of Shanghai Grant 12XD1406800 & 12JC1409000, and CAS/SAFEA International Partnership Program for Creative Research Teams are acknowledged.

References

  1. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 69, 183 CrossRef PubMed.
  2. K. Geim, Science, 2009, 324, 1530 CrossRef PubMed.
  3. S. Bai and X. Shen, RSC Adv., 2012, 2, 64 RSC.
  4. X. Huang, X. Qi and F. Boey, Chem. Soc. Rev., 2012, 41, 666 RSC.
  5. B. Kumar, K. Y. Lee, H. Park, S. J. Chae, Y. H. Lee and S. Kim, ACS Nano, 2011, 5, 4197 CrossRef CAS PubMed.
  6. H. Wang, H. S. Casalongue, Y. Liang and H. Dai, J. Am. Chem. Soc., 2010, 132, 7472 CrossRef CAS PubMed.
  7. J. F. Liang, W. Wei, D. Zhong, Q. L. Yang, L. D. Li and L. Guo, ACS Appl. Mater. Interfaces, 2012, 4, 454 CAS.
  8. K. Zhang, L. L. Zhang, X. S. Zhao and J. Wu, Chem. Mater., 2010, 22, 1392 CrossRef CAS.
  9. J. Shen, B. Yan, M. Shi, H. Ma, N. Li and M. Ye, J. Colloid Interface Sci., 2011, 356, 543 CrossRef CAS PubMed.
  10. Y. G. Guo, Y. S. Hu, W. Sigle and J. Maier, Adv. Mater., 2007, 19, 2087 CrossRef CAS.
  11. G. Sudant, E. Baudrin, D. Larcher and J. Tarascon, J. Mater. Chem., 2005, 15, 1263 CAS.
  12. L. J. Fu, T. Zhang, Q. Cao, H. P. Zhang and Y. P. Wu, Electrochem. Commun., 2007, 9, 2140 CrossRef CAS.
  13. S. Ding, J. S. Chen, D. Luan, F. Y. C. Boey, S. Madhavibc and X. W. Lou, Chem. Commun., 2011, 47, 5780 RSC.
  14. H. Q. Cao, B. J. Li, J. X. Zhang, F. Lian, X. H. Kong and M. Z. Qu, J. Mater. Chem., 2012, 22, 9759 RSC.
  15. L. He, R. Ma, N. Du, J. Ren, T. Wong, Y. Li and S. T. Lee, J. Mater. Chem., 2012, 22, 19061 RSC.
  16. Y. Y. Liang, H. L. Wang, H. S. Casalongue, Z. Chen and H. J. Dai, Nano Res., 2010, 3, 701 CrossRef CAS.
  17. Y. Tang, L. Yang, J. Chen and Z. Qiu, Langmuir, 2010, 26, 10111 CrossRef CAS PubMed.
  18. L. Shen, C. Yuan, H. Luo, X. Zhang, S. Yang and X. Lu, Nanoscale, 2011, 3, 572 RSC.
  19. Z. Wang, D. Xu, Y. Huang, Z. Wu, L. Wang and X. Zhang, Chem. Commun., 2012, 48, 976 RSC.
  20. J. Zhou, H. Song, L. Ma and X. Chen, RSC Adv., 2011, 1, 782 RSC.
  21. J. Jiang, C. G. Tian, W. Zhou, J. Q. Wang, Y. Xie, Q. J. Pan, Z. Y. Ren, Y. Z. Dong, D. Fu, J. L. Han and H. G. Fu, Chem.–Eur. J., 2011, 17, 8379 CrossRef PubMed.
  22. S. Yang, X. Feng, S. Ivanovici and K. Müllen, Angew. Chem., Int. Ed., 2010, 49, 8408 CrossRef CAS PubMed.
  23. X. Zhang, P. S. Kumar, V. Aravindan, H. H. Liu, J. Sundaramurthy, S. G. Mhaisalkar, H. M. Duong, S. Ramakrishna and S. Madhavi, J. Phys. Chem. C, 2012, 116, 14780 CAS.
  24. S. Yang, X. Feng and K. Müllen, Adv. Mater., 2011, 23, 3575 CrossRef CAS PubMed.
  25. D. Cai, D. Li, S. Wang, X. Zhu, W. Yang, S. Zhang and H. Wang, J. Alloys Compd., 2013, 561, 54 CrossRef CAS.
  26. G. F. Ortiz, I. Hanzu, P. Knauth, P. Lavela, J. L. Tirado and T. Djenizian, Electrochim. Acta, 2009, 54, 4262 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available: XRD, TEM, SEM and EDS analysis of samples. See DOI: 10.1039/c4ra05027d

This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.