One-pot synthesis of 3-D dandelion-like architectures constructed by rutile TiO₂ nanorods grown along [001] axis for high-rate lithium ion batteries

Abstract

Novel dandelion-like rutile TiO₂ superstructures are synthesized through a facile one-pot hydrolysis route. The as-prepared structures are composed of inter-aggregated straight nanorods, which are constructed from nanosized rutile TiO₂ (∼6 nm) grown along the [001] direction. The as-derived TiO₂ shows a high reversible lithium storage capacity of 242 mA h g⁻¹ and an excellent rate capability of 116 mA h g⁻¹ at 20 C.

---

Introduction

In recent years, lithium ion batteries (LIBs) have attracted considerable attention as a leading candidate for hybrid electric vehicles (HEVs) and electric vehicles (EVs) in view of the gradual depletion of oil resources and global warming.¹ However, the traditional carbonaceous anodes of commercial LIBs cannot satisfactorily meet the safety requirements for future applications in HEVs and EVs. To solve this problem, many efforts have focused on exploiting alternative anode materials. TiO₂ has been regarded as a promising anode material for high power LIBs^2,3 owing to its inherent advantages. Compared to graphite, the TiO₂ shows higher lithium insertion potential (1–3 V vs. Li/Li⁺), which could prevent the formation of dendritic lithium during the charge–discharge process. Moreover, the TiO₂ features negligible volume change (<1%) during the charge–discharge process, which could ensure an excellent reversibility and structure stability in a long-term cycle. TiO₂ polymorphs include rutile, anatase, brookite, and TiO₂-B. Up to now, the investigations have mainly focused on anatase⁴ and TiO₂-B.⁵ Although rutile (P42/mnm) is the most common and thermodynamically stable form of TiO₂, there are few reports on Li⁺ insertion properties of rutile TiO₂. The primary reason is that the bulk rutile can only accommodate a negligible amount of Li⁺ (<0.1 Li⁺ per TiO₂ unit) due to kinetic restrictions.^3a,6

To overcome this issue, rutile TiO₂ has been prepared as nanosized or mesoporous particles to increase the electrode/electrolyte contact area and shorten the Li⁺ diffusion length in the solid phase, facilitating for Li⁺ insertion/extraction.⁷ Recently, the reversible capacity of rutile TiO₂ was greatly increased to 200 mA h g⁻¹ (∼0.6 Li/Ti),^7a,d which is even superior to that of most nanosized anatase (168 mA h g⁻¹, ∼0.5 Li/Ti). However, the capacities of the as-reported nanosized rutile TiO₂ are still lower than its theoretical capacity value (∼0.85 Li/Ti),⁸ since the diffusion of Li⁺ in rutile is highly anisotropic in the tetragonal rutile (P42/mnm) special arrangement. Experimental results and simulations have revealed that the Li⁺ diffusion coefficient along the c-direction is approximately 10⁻⁶ cm² s⁻¹, which is much higher than that in the ab-plane (10⁻¹⁵ cm² s⁻¹).^3a,6a Therefore, the migration of Li⁺ in rutile TiO₂ is nearly confined along the c-axis channels. In this context, constructing superstructures composed of nanosized rutile TiO₂ grown along c-axis which makes all the Li⁺ diffusion channels available from the surface is of great significance for enhancing its electrochemical performance.

Herein, we report the synthesis of “dandelion” rutile TiO₂ through a facile one-pot hydrolysis route by employing titanium tetrachloride (TiCl₄) as the titanium source and 1-hexadecyl-3-methyl imidazolium bromine (C16mimBr) as the structure-directing agent. The TiO₂ nanostructure is composed of inter-aggregated straight nanorods grown along c-axis, which facilitates the transport of lithium ions and electrons. The as-prepared rutile TiO₂ exhibits a high reversible capacity of 242 mA h g⁻¹ (∼0.72 Li/Ti) and a high rate capacity of 116 mA h g⁻¹ at 20 C with an excellent cycle life.

Experimental

Synthesis of rutile TiO₂

In a typical synthesis process of hierarchical nanostructured rutile TiO₂, we employ TiCl₄ as precursor, and C16mimBr as solvent and structure-directing agent. The chemical structure of C16mimBr is shown schematically below.

1.1 mL of TiCl₄ was dropwise added into 5.4 mL of distilled water mixed with 3.23 g of C16mimBr in an ice bath, stirring strongly to form a clear solution. Thereafter, in order to achieve highly ordered superstructures, the solution was heated to 100 °C for another 12 h under strong stirring to control the hydrolysis of TiCl₄ in the solution. The obtained dispersion was diluted with 20 mL of anhydrous ethanol, then gathering the product by centrifugation. The residual of C16mimBr in the product was removed by extracting the sample with acetonitrile at 100 °C. We carry out Fourier transform infrared (FTIR) measurement to see if there is some residual C16mimBr in our sample. As shown in Fig. S1 and S2,† the imidazolium ν(C–H) stretching region (3200–3000 cm⁻¹) of the C16mimBr disappears, indicating that the C16mimBr is completely removed.

Physics characterisation

The resulting samples were characterized by means of X-ray diffraction (XRD, Rigaku D/max-γB) with monochromated Cu Kα radiation at a scanning rate of 2° min⁻¹ in the range of 10–70°. The morphology of the synthesized materials was examined using a scanning electron microscopy (SEM, Hitachi S4800) and the microstructure of the powders was observed by a high resolution transmission electron microscopy (HRTEM, Hitachi 7650) operating at 300 kV. Before TEM and HRTEM characterizations, the TiO₂ sample was dried and then dispersed thoroughly in ethanol by ultrasonic method. The as-obtained suspension was dropped onto a 400-mesh carbon-coated copper grid and left in air to dry. The surface area for powdered sample was measured by the Brunauer–Emmett–Teller (BET) method using ASAP2020.

Cell fabrication and electrochemical characterizations

The anode film studied here was prepared by mixing the TiO₂ powder, carbon black and polyvinylidene fluoride with a weight ratio of 8 : 1 : 1 in N-methyl pyrrolidinone. The slurry was coated onto an aluminum foil by the “doctor blade” technique and dried under vacuum at 120 °C over night. The electrode was roll-pressed under an appropriate pressure to enhance the adhesion. The anodes typically had an active material loading of 2.4–3.2 mg and their loading differences for the same electrochemical test were less than 0.15 mg. The compact density and thickness of anodes were 0.75 g cm⁻³ and 20–30 μm, respectively. Coin type (CR2025) test cells were assembled in an Mbraun argon-filled glove box (O₂ < 1 ppm and H₂O < 1 ppm) using two porous polypropylene films as a separator, 1 M LiPF₆ in ethylene carbonate, dimethyl carbonate and ethylmethyl carbonate (EC/DMC/EMC, 1 : 1 : 1 vol) as electrolyte, and Li foil as the counter and reference electrodes. Constant current charge–discharge was performed at various rates within a voltage window of 1–2.5 V (vs. Li/Li⁺). CV measurements were carried out on an electrochemical workstation (CHI, 650B) at a series of sweep rates. Electrical impedance spectroscopy (EIS) experiments were carried out on a Parstat 2273 advanced electrochemical system in the frequency range mainly from 1 MHz to 50 mHz with an amplitude of 10 mV. Before EIS measurements all samples were charged to the same voltage of 1 V.

Results and discussion

The detailed synthesis procedure of rutile TiO₂ is described in Experimental section. The crystal phase of the as-derived TiO₂ was determined by the X-ray diffraction (XRD) characterization. As shown in Fig. 1a, all diffraction peaks can be indexed to TiO₂ crystal with the standard tetragonal rutile pattern (JCPDS no. 65-0191) and no secondary peak is observed over the entire pattern. Meanwhile, the average crystal size is calculated to be 6.2 nm from the most intense diffraction peak (110) through the Debye–Scherer equation, indicating that the product is composed of pure rutile TiO₂ nanocrystals.

Fig. 1 (a) XRD of standard rutile TiO₂ (JCPDS no. 65-0191) and the as-derived TiO₂. (b) Representative SEM image of rutile TiO₂. (c) TEM images of dandelion-like rutile TiO₂ (the inset shows the TEM image of a TiO₂ particle). (d) HRTEM image of rutile TiO₂ (the inset shows the corresponding FFT pattern of the same region).

The panoramic scanning electron microscopy (SEM) image in Fig. 1b shows the existence of uniformly distributed spherical-shaped particles with an average size of 400 nm. The discrepancy of the sizes calculated from XRD and observed in SEM measurements suggests that the as-prepared TiO₂ particles are consist of nanosized subunits.

We further carried out transmission electron microscopy (TEM) measurements to elucidate the intrinsic micro/nano structure of the rutile TiO₂ particles. The TEM image (the inset of Fig. 1c) reveals that the rutile TiO₂ particle features a dandelions-like nanostructure comprised with numerous well-defined and straight nanorods. As shown by the magnified TEM image (Fig. 1c), the nanorods with a diameter of around 6 nm are oriented radially from the central region toward edges of the particle. In addition, the result of the nitrogen adsorption–desorption experiment (Fig. S3 in ESI†) indicates that there exists a large BET specific surface area of 114 m² g⁻¹ for the dandelions-like TiO₂. Fig. 1d shows the high-resolution transmission electron microscopy (HRTEM) image and the corresponding fast Fourier-transform (FFT) pattern of the selected region as marked in Fig. 1c. The crystalline region with clear lattice fringes has an inter-planar spacing of 0.32 nm, which is consistent with the (110) atomic planes of the rutile structure. It indicates that rutile nanorods are single crystalline along the [001] axis. The diffusion ring in the inset of Fig. 1d stems from the amorphous carbon substrate. When we remove the C16mimBr in the synthesis procedure, the nanocrystalline anatase TiO₂ with an average nanoparticles size of ∼4 nm is obtained (Fig. S4 and S5 in ESI†). Therefore, the C16mimBr plays a critical role in controlling the crystal structure and the morphology of TiO₂. As reported by Chang,⁹ the hydrogen atoms in C2 position of imidazole rings have strong hydrogen-bonding interaction with oxygen atoms of the rutile surface. The [C16mim]⁺ cations can effectively anchor onto the (110) facets of TiO₆ octahedra instead of H⁺, forming [C16mim]–O–Ti hydrogen-bonding groups as illustrated in Fig. 2a. The mutual π–π stacking interactions between aromatic rings would induce the formation of linear nuclei via edge-sharing TiO₆ octahedra. Meanwhile, the hydrophobic interaction between long alkyl chains can synergetically extend the π–π stacking effect and thus increase the long-range oriented-growth of rutile TiO₂.¹⁰ Therefore, the rutile nanorods grown along [001] axis with a diameter of about 6 nm can be constructed. The rutile nanorods were further assembled into dandelions-like nanoarchitecture, owing to the hydrophobic interactions between the adjacent hydrocarbon chains enwrapped on the surface of the nanorods.

Fig. 2 (a) Schematic illustration of a projected view of the nucleation pathways by [C16mim]⁺ ions anchored onto rutile (110) plane. (b) The schematic model of rutile TiO₂ with insertion of Li⁺ ions between TiO₆ octahedral voids.

To evaluate the electrochemical characterizations of the as-prepared dandelions-like rutile TiO₂, we first measured its cyclic voltammogram (CV) between 1 and 3 V as shown in Fig. 3a. At the first discharge, there are three well-resolved cathodic peaks centered at 1.1, 1.4, and 2.1 V, which disappear in the following cycles. The peaks at 1.1 and 1.4 V can be ascribed to the phase transformation from TiO₂ to Li_xTiO₂, and the peak at 2.1 V is assigned to the irreversible adsorption of lithium ions.^7a In the subsequent cycles, a pair of broad cathodic/anodic peaks centred at around 1.76 and 1.83 V are associated with lithium insertion/extraction in the Li_xTiO₂. Moreover, the difference between the second cycle and the subsequent cycles is negligible, revealing that the dandelions-like rutile TiO₂ displays a good reversible capacity.

Fig. 3 (a) Cyclic voltammograms of rutile TiO₂ electrode at a scan rate of 0.2 mV s⁻¹ for the first 10 cycles. Potentials are measured relative to a Li/Li⁺ reference electrode (potential range: 1–3 V). (b) Voltage profiles of initial charge–discharge cycles of Li/TiO₂ between 1 V and 2.5 V at a rate of 0.2 C (= 68 mA g⁻¹). (c) Discharge capacity and coulombic efficiency versus cycle number under different current rates (1–20 C).

The lithium insertion/extraction properties could be further corroborated by the charge–discharge measurement at a current density of 68 mA g⁻¹ (rate = 0.2 C) as shown in Fig. 3b. In the first cycle, we observe an irreversible capacity with a discharge capacity of 337 mA h g⁻¹ and a charge capacity of 269 mA h g⁻¹. This irreversible capacity is comparable with the values reported by other groups and is normally attributed to the formation of intermediate phase Li_xTiO₂ and the irreversible surface adsorption of lithium ions.⁷ During subsequent cycles, the profiles feature almost monotonic voltage evolution without a constant potential region, reflecting the typical behavior of Li⁺ insertion/extraction in the Li_xTiO₂. The second charge–discharge curves indicate that about 0.72 Li per mol TiO₂ can be reversibly inserted/extracted, and the corresponding capacity of 242 mA h g⁻¹ is much higher than that reported previously.^6,7 In addition, between the second and subsequent cycles, no obvious difference can be observed from the charge–discharge capacities, in agreement with our aforementioned CV results. This behavior could be explained by the fact that rutile TiO₂ nanorods have freely accessible parallel channels along the [001] direction as illustrated in Fig. 2b, in which lithium ions can be accommodated without causing any remarkable distortion of the prepared superstructure.

The rate capability of the as-derived rutile TiO₂ was further evaluated with charge–discharge rates stepwise increasing from 1 to 20 C. For each stage, the cell was cycled for 50 times. As shown in Fig. 3c, the discharge capacities at 1 C, 5 C, 10 C and 20 C are around 188, 170, 144 and 116 mA h g⁻¹, respectively. The rate performance of our TiO₂ is superior to that of rutile-TiO₂-based nanomaterials under similar test conditions in previous reports.^6,7 We ascribe the enhanced rate capability to the following reasons: (i) for the nanosized TiO₂, the pseudocapacitance contributes obviously to the electrochemical lithium storage and could significantly accelerate the charge–discharge kinetics of the material;¹¹ (ii) the well-connected nanorod-crystals provide a continuous pathway for the diffusion of lithium ions and electrons in the titania scaffold; (iii) the special structure exposes plenty of vertical cross-section of c-axis as illustrated in Fig. 2b, thus enlarging the effective contact area for the transport of Li⁺.

In addition, the rutile TiO₂ exhibits excellent cyclability, retaining over 97% of its initial capacity after 50 cycles at all rates. Impedance analysis was further carried out after various numbers of cycles (including 1^st, 50^th, 100^th, 200^th). As shown in Fig. 4, the EIS spectra consists of a semicircle and a slope, referring to the charge transfer reaction and the diffusion of Li⁺ in the bulk electrode, respectively.¹² During our experiment, the values of charge-transfer resistance (R_ct) are all lower than 70 Ω, implying negligible structure change of dandelions-like rutile TiO₂ during cycling process. The SEM and TEM measurements for the TiO₂ anode after 200 cycles (Fig. S6†) further confirms little structure change of the TiO₂ nanorods after the cycling process.

Fig. 4 EIS data at different stages.

Furthermore, the coulombic efficiencies approach 100% even at higher identical charge and discharge rates (right ordinate of Fig. 3c). The good electrochemical performance in combination with a facile preparation procedure makes the described dandelions-like rutile TiO₂ a promising electrode material for lithium ion batteries.

To summarize, we have synthesized a dandelions-like rutile TiO₂ superstructure through a facile one-pot hydrolysis route by employing titanium tetrachloride as the titanium source and C16mimBr as the structure-directing agent. The dandelion-like particles with the average size of 400 nm are comprised of inter-aggregated straight rods, which constructed from nanosized rutile TiO₂ (∼6 nm) grown along [001]. The cell assembled with dandelions-like rutile TiO₂ shows high reversible lithium storage capacity and excellent rate capability. The strategy of constructing dandelions-like superstructures demonstrated in this work may be potentially extendable to other lithium insertion host materials for future improvement of their specific capacities and rate performance.

Acknowledgements

This project is supported by the National Natural Science Foundation of China (no. 51203036), the China Postdoctoral Science Special Foundation (2013T60380), the China Postdoctoral Science Foundation (2012M520748), and the “Young Talent Program” of Harbin Institute of Technology. We are grateful to Lanzhou Greenchem ILS, LICP. CAS. China for supplying C16mimBr. The authors also thank Dr Yunfeng Qiu from Academy of Fundamental and Interdisciplinary Sciences, Harbin Institute of Technology, for his constructive suggestions.

Notes and references

1. (a) M. Armand and J. M. Tarascon, Nature, 2008, 451, 652; (b) P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930.
2. G. Nuspl, K. Yoshizawa and T. Yamabe, J. Mater. Chem., 1997, 7, 2529.
3. (a) D. Deng, M. G. Kim, J. Y. Lee and J. Cho, Energy Environ. Sci., 2009, 2, 818; (b) T. Djenizian, I. Hanzu and P. Knauth, J. Mater. Chem., 2011, 21, 9925; (c) G. Jeong, Y.-U. Kim, H. Kim, Y.-J. Kim and H.-J. Sohn, Energy Environ. Sci., 2011, 4, 1986; (d) A. I. Hochbaum and P. Yang, Chem. Rev., 2010, 110, 527; (e) Z. Yang, D. Choi, S. Kerisit, K. M. Rosso, D. Wang, J. Zhang, G. Graff and J. Liu, J. Power Sources, 2009, 192, 588; (f) P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930.
4. (a) M. Wagemaker, A. P. M. Kentgens and F. M. Mulder, Nature, 2002, 418, 397; (b) K. X. Wang, M. D. Wei, M. A. Morris, H. S. Zhou and J. D. Holmes, Adv. Mater., 2007, 19, 3016; (c) S.-M. Paek, J.-H. Kang, H. Jung, S.-J. Hwanga and J.-H. Choy, Chem. Commun., 2009, 48, 7536; (d) S. Liu, G. L. Pan, N. F. Yan and X. P. Gao, Energy Environ. Sci., 2010, 3, 1732; (e) J. S. Chen, Y. L. Tan, C. M. Li, Y. L. Cheah, D. Y. Luan, S. Madhavi, F. Y. C. Boey, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2010, 132, 6124; (f) K. Saravanan, K. Ananthanarayanan and P. Balaya, Energy Environ. Sci., 2010, 3, 939; (g) C. H. Sun, X. H. Yang, J. S. Chen, Z. Li, X. W. Lou, C. Li, S. C. Smith, G. Q. Lu and H. G. Yang, Chem. Commun., 2010, 46, 6129; (h) J. S. Chen, D. Luan, C. M. Li, F. Y. C. Boey, S. Qiao and X. W. Lou, Chem. Commun., 2010, 46, 8252; (i) S. Ding, J. S. Chen, D. Luan, F. Y. C. Boey, S. Madhavi and X. W. Lou, Chem. Commun., 2011, 47, 5780.
5. (a) A. R. Armstrong, G. Armstrong, J. Canales and P. G. Bruce, Angew. Chem., Int. Ed., 2004, 43, 2286; (b) G. Armstrong, A. R. Armstrong, J. Canales and P. G. Bruce, Chem. Commun., 2005, 2454; (c) J. Li, W. Wan, H. Zhou, J. Li and D. Xu, Chem. Commun., 2011, 47, 3439; (d) A. R. Armstrong, C. Arrouvel, V. Gentili, S. C. Parker, M. S. Islam and P. G. Bruce, Chem. Mater., 2010, 22, 6426; (e) H. Kaper, S. Sallard, I. Djerdj, M. Antonietti and B. M. Smarsly, Chem. Mater., 2010, 22, 3502.
6. (a) W. J. Macklin and R. J. Neat, Solid State Ionics, 2003, 157, 35; (b) Y.-S. Hu, L. Kienle, Y.-G. Guo and J. Maier, Adv. Mater., 2006, 18, 1421.
7. (a) E. Baudrin, S. Cassaignon, M. Koesch, J. P. Jolivet, L. Dupont and J. M. Tarascon, Electrochem. Commun., 2007, 9, 337; (b) C. H. Jiang, I. Honma, T. Kudo and H. S. Zhou, Electrochem. Solid-State Lett., 2007, 10, A127; (c) P. Kubiak, M. Pfanzelt, J. Geserick, U. Homannc, N. Husing, U. Kaiserc and M. Wohlfahrt-Mehrensa, J. Power Sources, 2009, 194, 1099; (d) M. Pfanzelt, P. Kubiak and M. Wohlfahrt-Mehrens, Electrochem. Solid-State Lett., 2010, 13, A91; (e) Y. Jin, Y. Liu, Y. Wang, X. Li, S. Hu and W. Li, Int. J. Miner., Metall. Mater., 2012, 19, 1058.
8. M. Wagemaker, W. J. H. Borghols, U. Lafont, E. M. Kelder and F. M. Mulder, Chem. Mater., 2008, 20, 2949.
9. J.-G. Chang, J. Wang and M. C. Lin, J. Phys. Chem. A, 2007, 111, 6746.
10. Y. Zhou, J. H. Schattka and M. Antonietti, Nano Lett., 2004, 4, 477.
11. (a) T. Brezesinski, J. Wang, J. Polleux, B. Dunn and S. H. Tolbert, J. Am. Chem. Soc., 2010, 131, 1802; (b) J. Wang, J. Polleux, J. Lim and B. Dunn, J. Phys. Chem. C, 2007, 111, 14925.
12. H. Liu, Y. Feng, K. Wang and J. Xie, J. Phys. Chem. Solids, 2008, 69, 2037.

---

Footnote

† Electronic supplementary information (ESI) available: Additional data analysis. See DOI: 10.1039/c5ra00006h

---