Facile synthesis of a mesostructured TiO₂–graphitized carbon (TiO₂–gC) composite through the hydrothermal process and its application as the anode of lithium ion batteries

Abstract

A mesostructured TiO₂–graphitic carbon (TiO₂–gC) composite was synthesized through a simple and scalable hydrothermal method to be employed as an anode material in Li-ion batteries. In a wide voltage range (0.0–2.5 V), the TiO₂–gC composite anode possesses a high initial lithiation capacity (598 mA h g⁻¹) at 0.1 C (1 C: 150 mA g⁻¹), and it still retains 369 mA h g⁻¹ after 50 cycles at 0.5 C. Furthermore, under a high current density of 2 C, the TiO₂–gC anode exhibits stable capacity (252 mA h g⁻¹) retention for up to 200 cycles. This excellent electrochemical performance could be ascribed to a synergistic effect of well-developed mesoporosity with a high surface area (345.4 m² g⁻¹), the conductive graphitic carbon wall, and uniformly dispersed TiO₂ nanoparticles, resulting in improved Li⁺ penetration, fast electron transport and high structural stability during cycling.

---

Introduction

Rechargeable Li-ion batteries (LIBs) have attracted considerable attention as promising electrochemical energy storage devices for electric vehicles due to their high energy density, long cycling life and environmental benignness.^1–4 Graphite, a typical LIB anode material, has been employed due to its high reliability. However, several shortcomings of graphite anodes including their limited discharge capacity (372 mA h g⁻¹), low current density, and some safety issues, have driven the development of new anode materials.^5,6 Among the various anode materials, titanium oxide (TiO₂) has received a lot of interest as an alternative anode material due to its superior safety, low volume change (<4% for Li_xTiO₂, 0 ≤ x ≤ 1), environmental friendliness, low cost, and high chemical stability.^7–9 However, the application of TiO₂ as a LIB anode material has been limited by its low electric conductivity (∼10⁻¹² to 10⁻⁷ S cm⁻¹) and sluggish Li⁺ transport (∼10⁻¹⁵ to 10⁻⁹ cm² s⁻¹), originating from its wide energy band gap (E_g = 3.0–3.4 eV), resulting in the drastic fading of capacity at high current density or after long cycling.^10–12

To address the aforementioned problems of TiO₂ as an anode material, several methods, including the fabrication of thin films,⁸ the preparation of composites with carbon^8–10 and nanostructure formation,^13–15 have been developed to regulate the conductivity, crystalline structure, and surface area of TiO₂.¹⁶ Among the above, it is known that the surface area and conductivity of TiO₂ play the most crucial roles in enhancing the lithium storage capability and reaction kinetics, since the ion diffusion pathway and electron transport in the electrode are largely dependent on them.^17–19

In this context, approaches to the formation of TiO₂-based porous carbon composites have attracted special attention since carbon materials can provide good electrical conductivity and have a large porosity and specific surface area.^20–22 Up to this date, carbon materials within the TiO₂-composite have hardly reached our requirements due to their low porosity and amorphous phase, leading to a relatively low electrical conductivity and sluggish Li⁺ transport through the TiO₂ phase during the lithiation–delithiation process.²³ For enhanced conductivity and high porosity, it is recommended to form mesostructured carbon–TiO₂ composites in a graphitic carbon phase.^7,9,11 To our best knowledge, most of the reported synthetic methods for TiO₂–C composites have failed to satisfy such requirements (high conductivity, porosity), as well not being viable for practical applications. A facile method to synthesize a mesostructured composite of TiO₂ and highly conductive porous carbon remains essential, but challenging.

In the present work, we report a simple and facile method for the synthesis of a mesostructured TiO₂–graphitic carbon (TiO₂–gC) composite via a hydrothermal approach. Using a Ni catalyst initially included in the composite precursor, a well-developed graphitic carbon–TiO₂ composite was obtained without further graphitization. Additionally, the strong interaction between TiO₂ and the carbon precursors can inhibit the growth of TiO₂ nanocrystals, exerting a positive effect on the homogeneous distribution of the as-formed TiO₂ nanoparticles. Notably, in contrast to conventional carbon composite syntheses requiring complicated steps, our process is simple, scalable and suitable for practical application. We analyzed the as-formed TiO₂–gC mesostructure by various morphological, structural and spectroscopic methods. Taking the advantages of our TiO₂–gC composite as an anode material in LIBs, we further investigated its electrochemical performance.

Experimental

Material synthesis

All materials were purchased from Sigma-Aldrich and used without further purification.

Porous TiO₂–graphitic carbon composite (TiO₂–gC). The TiO₂–gC composite was prepared by a hydrothermal synthesis followed by a sintering process, as illustrated in Fig. 1. The hydrothermal synthesis was performed using tetrabutyl titanate, glucose and nickel nitrate as the Ti source, carbon precursor and Ni catalyst, respectively. In a typical synthesis, 8.4 g of glucose, 8.4 g of titanium isopropoxide, 5 g of nickel nitrate and 0.5 g of HCl solution (37%) were mixed and dissolved in 40 mL of water under magnetic stirring until completely homogenized. The mixture was transferred to a hydrothermal bomb and kept still at 180 °C for 18 hours. Afterwards, the as-prepared hydrothermal product was sintered at 900 °C for 2 h under an argon atmosphere with a heating rate of 5 °C min⁻¹. The resulting composite was treated with 2 M HCl solution to obtain a TiO₂–gC composite with a pure graphitic carbon framework by removing the nickel content in the composite. Graphitic carbon (gC) was prepared by the same procedure without adding titanium isopropoxide.

Material characterization

The size and morphology of the samples were analyzed by transmission electron microscopy (TEM) and scanning electron microscopy (SEM). Regular TEM images and high-resolution TEM (HR-TEM) images were obtained using an FEI CM120 and a JEM-2100F field emission microscope at accelerating voltages of 100 kV and 200 kV, respectively. Scanning TEM (S-TEM) and elemental maps were carried out under high-angle annular dark field (HAADF) mode using the HR-TEM. SEM images were obtained with a NOVA NanoSEM 630 apparatus at an accelerating voltage of 1.5 kV. The crystalline structures of the samples were characterized using X-ray diffractometry (XRD) using RAD-3C (Rigaku) with CuKa radiation (λ = 1.541 Å). The chemical bonds and elemental compositions of the samples were analyzed via X-ray photoelectron spectroscopy (XPS, Escalab 250, Thermo Fisher Sci.) with an AlKα (h_m: 1486.6 eV) source and inductively coupled plasma mass spectrometry (ICP-MS, Aurora M90, Bruker). The carbon content of the composite was analyzed by thermogravimetric analysis (TGA) (TGA conditions: from 25 °C to 800 °C/atmosphere (air), ramping rate: 2 °C min). Raman spectra of the samples were obtained with a Renishaw Raman Spectrometer under backscattering geometry. The spectra were averaged over 20 accumulations and the laser power was kept at 2.5 mW. N₂ sorption isotherms for the materials were performed at −196 °C with a porosity analyzer (Micromeritics ASAP 2000). The specific surface area of the material was calculated by the Brunauer–Emmett–Teller (BET) method. The pore volumes were obtained from the desorption branches of the isotherms using the Barrett–Joyner–Halenda (BJH) method.

Electrochemical analysis

The electrochemical properties of the samples were evaluated via 2032-type coin half-cells with sample electrodes as working electrodes and lithium foil as a counter electrode. The working electrode was composed of 80 wt% active material, 10 wt% super-P as a conductive additive, 10 wt% polyvinylidene fluoride (PVDF, KF1300) and an oxalic acid binder (0.02%) dissolved in N-methyl-2-pyrrolidinone. The working electrodes were prepared with a sample of ca. 1.0 g cm⁻³ on the copper current collector. The coin cells were assembled in an argon-filled glove box with an electrolyte of 1.0 M LiPF₆ in ethylene carbonate (EC)/ethyl methyl carbonate (EMC) (3 : 7 volume ratio). A galvanostatic lithiation/delithiation cycling test was performed in the potential range of 0.0–2.5 V using the WBCS3000 system (Wonatech Co., Ltd.) at 0.1C for the first two cycles, and a higher current density for the subsequent cycles. Electrochemical impedance spectroscopy (EIS) was measured in the frequency range of 100 kHz to 0.01 Hz. Cyclic voltammogram (CV) measurements were performed using Iviumstat (Ivium Technologies) over 0.0 to 2.5 V vs. Li/Li⁺ at a scan rate of 0.1 mV s⁻¹. All electrochemical tests were carried out at room temperature.

Results and discussion

Material characteristics

The porous spherical TiO₂–gC composite was prepared by a hydrothermal process followed by a subsequent sintering process. Typically, ∼5 g of the TiO₂–gC composite was produced per batch. This could be scaled-up through the regulation of the reactor volume and its parallel production. Fig. 1 illustrates the synthetic scheme of the TiO₂–gC composite from precursor to product through hydrothermal synthesis (a and b), calcination (b and c) and nickel removal (c and d). Fig. 1a shows a precursor containing Ti isopropoxide, glucose and nickel nitrate. Fig. 1b exhibits the as-formed composite of polymerized glucose–TiO₂ and nickel after a hydrothermal reaction at 180 °C for 18 hours, where TiO₂ and Ni(OH)₂ nanoparticles are distributed throughout the glucose polymer matrix. As shown in Fig. 1c and d, the subsequent processes of calcination of (b) under an argon atmosphere at 900 °C (Fig. 1c), and the removal of nickel from (c) (Fig. 1d) allows the formation of a robust TiO₂–gC composite. Note that our method is highly competitive to produce high performance electrode materials via a simple and scalable process using inexpensive precursors.

Fig. 1 Synthetic scheme of TiO₂–graphitic carbon (TiO₂–gC) composite: (a) precursor solution of Ti, glucose and nickel salt, (b) TiO₂–glucose–Ni composite after hydrothermal synthesis, (c) TiO₂–gC–Ni composite after calcination at 900 °C under an argon atmosphere, (d) TiO₂–gC composite after the removal of Ni.

Fig. 2 and S1† displays electronic images of the mesostructured TiO₂–gC composite, to investigate its morphology by scanning electron microscopy (SEM) in Fig. 2a and S1† and transmission electron microscopy (TEM) in Fig. 2b–e and S2,† respectively. First, as shown in Fig. 2a and S1a,† the composite particles are formed by the aggregation of small primary particles, exhibiting a polysized distribution ranging from 50 to 500 nm. The inset shows a digital photo-image of the TiO₂–gC composite in the form of a black colored powder. The TEM images (Fig. 2b, c and S2†) of the TiO₂–gC composite clearly reveal that the TiO₂ nanocrystal cores (dashed circles, average particle size: ca. 8.2 nm) are homogenously embedded into the framework of the porous graphitic carbon shell, indicating a uniform distribution of ultrafine TiO₂ nanocrystals in the mesostructured carbon matrix. In addition, Fig. S2† exhibits that the as-formed pores in the composite particles have a broad size distribution of 3–30 nm. Such uniform distribution of TiO₂ nanoparticles in the carbon matrix can be attributed to the unique formation process of the graphitic carbon layers via the catalytic carbonization of the carbon precursor (glucose) by nickel nanoparticles, which effectively prohibit the aggregation of TiO₂ nanoparticles.^11,24 More importantly, as graphitic carbon efficiently facilitates electron transport, the electrical conductivity of the composite can be greatly improved. The high-magnification TEM (HR-TEM) images (Fig. 2c and d) also confirm the composite structure of the TiO₂ nanoparticles (core) covered by the graphitic carbon (shell). We found that the TiO₂ nanocrystal (Fig. 2c) consists of (110) facets with a d-spacing of 0.136 nm, while the lattice fringe of graphitic carbon with a d-spacing of 0.342 nm is observed, which is similar to the (002) facet of typical graphite with a d-spacing of 0.335 nm (Fig. 2d).^11,24 Fig. 2e exhibits elemental mapping of TiO₂–gC, confirming uniformly distributed elements (C, Ti, O), consistent with the SEM elemental mapping result (Fig. S1b†). We further analyzed the composition with thermogravimetric analysis (TGA) (Fig. S3†) and inductively coupled plasma mass spectrometry (ICP-MS). The elemental composition of the TiO₂–gC composite in mass ratio was estimated to be Ti (36.1%), O (24.1%) and C (39.8%) (TiO₂: 60.2%, graphitic carbon: 39.8%). Note that the content of the TiO₂–gC composite before Ni removal was determined to be Ti (30.4%), O (19.8%), C (32.7%) and Ni (17.1%). We believe that the considerable amount of Ni nanoparticles embedded in the composite plays a crucial role as a catalyst to make graphitic carbon, thereby enhancing the conductivity of the TiO₂–gC composite.^11,24 Overall, the TiO₂–gC composite has a structure of TiO₂ nanoparticle cores embedded in a graphitic carbon layer shell, which could be conducive to enhancing its EC performance as a LIB anode.

Fig. 2 Morphology of the TiO₂–gC composite observed by electronic microscopy: (a) SEM image (inset: digital photo-image) (b) TEM image; high resolution TEM images for (c) TiO₂ nanoparticles embedded in TiO₂–gC, (d) graphitic carbon in TiO₂–gC, (e) elemental mapping of TiO₂–gC composite (all, carbon (green), titanium (violet), oxygen (blue)).

Fig. 3 presents analysis of the structure of the TiO₂–gC composite and controls with XRD (Fig. 3a and S4†), N₂ sorption (Fig. 3b and S5†), Raman spectra (Fig. 3c) and XPS analysis (Fig. 3c and d). Fig. 3a presents wide angle X-ray diffraction patterns (XRD) for TiO₂, gC and the TiO₂–gC composite (before and after sintering), as obtained from the different preparation stages. As displayed in the XRD pattern (Fig. 3a and S4†) for the TiO₂ nanocrystal and TiO₂–gC (before and after sintering), the TiO₂–gC composite has a well-defined crystalline structure with a major phase of anatase TiO₂ (A).^6,7,15–17 Before sintering, the TiO₂–gC composite exhibits characteristic XRD peaks of β-Ni(OH)₂ (Ni–OH) and TiO₂ (A), indicating the formation of Ni(OH)₂ from nickel nitrate and TiO₂ from Ti-isopropoxide through hydrothermal synthesis.²⁵ After sintering, as shown in Fig. S4,† most of the XRD peaks of TiO₂–gC can be assigned to rutile (R) and anatase (A) phases of crystalline TiO₂, while Ni(OH)₂ was transformed to Ni due to the carbothermal reduction.^{6,7,11,16,24,26} In addition, the peak that appeared around 25° might correspond to graphitic carbon at (002) orientation; but such peak assignment is not very clear because of the overlapping peak position with the anatase phase TiO₂ (101).^6,7,16,17,26 The co-existence of two phases (anatase and rutile) indicates the partial phase transformation from anatase to rutile by a sintering process, where a significant portion of the anatase TiO₂ is still retained even after the high temperature (900 °C) sintering process. This partial phase transformation is attributable to the existence of graphitic carbon from glucose which can serve as a binder to glue TiO₂ nanocrystals and limit its crystalline growth during the sintering process.²⁷ Note that the lack of significantly intensified diffraction peaks and crystalline size (4.5 nm (before) → 6.7 nm (after)) of the TiO₂ nanocrystals after the sintering process further suggests that the carbonized glucose can effectively suppress the crystalline growth of TiO₂.²⁷ Fig. 3b and S5† analyzes the porous structure of the TiO₂–gC composite and control samples (TiO₂ and gC) by measuring their N₂ sorption behavior and pore size distribution. The N₂ sorption curve (Fig. 3b) of the TiO₂–gC composite (BET surface area: 345.4 m² g⁻¹) displays type IV isotherm behavior with a capillary condensation in the relative pressure (P/P₀) range of 0.8–0.9.⁵ The inset plot of Fig. 3b shows that TiO₂–gC has a broad pore size distribution (1–50 nm) with an average pore size of ca. 13.8 nm and a large pore volume of 0.769 cm³ g⁻¹, consistent with the TEM results (Fig. S2†). Fig. S5† compares the N₂ sorption (Fig. S5a†) and pore size distribution (Fig. S5b†) for TiO₂–gC and the controls. TiO₂–gC has the largest surface area (SA: 345.4 m² g⁻¹) and pore volume (PV: 0.769 cm³ g⁻¹) compared with the controls (TiO₂: 5.4 m² g⁻¹ (SA), 0.011 cm³ g⁻¹ (PV), gC: 223.3 m² g⁻¹ (SA), 0.028 cm³ g⁻¹ (PV)). Moreover, gC and TiO₂ have a predominantly microporous structure with a smaller pore size (TiO₂: 11.4 nm, gC: 8.4 nm) compared to TiO₂–gC (13.8 nm).

Fig. 3 (a) Crystalline structure analysis by X-ray diffraction of TiO₂–gC composites and TiO₂; (b) N₂ sorption curve of TiO₂–gC composite (inset: pore size distribution); (c) Raman spectra of TiO₂–gC composite and TiO₂; XPS spectra of TiO₂–gC composite and TiO₂: (d) carbon (C1s), (e) titanium (Ti2p), and (f) oxygen (O1s).

Fig. 3c and S6† present Raman spectra of TiO₂–gC (before and after sintering) to describe the changes to the nature of the carbon in the composite. The spectrum of the TiO₂–gC displays two carbon related signature peaks positioned around 1349–1356 (D-band) and 1583–1596 cm⁻¹ (G-band).^6,7,11 The G-band for both TiO₂–gC composites (before and after sintering) confirms the presence and formation of sp²-hybridized graphitic carbon structures within the carbonaceous layers.^6,7,11 The D-band at 1341 cm⁻¹ indicates the presence of sp³-hybridized carbon and defects within the graphitic structure.^6,7,11 In addition, the enhanced intensity of the G-band of TiO₂–gC (after sintering) suggests the formation of more graphitic carbon.¹¹ The several peaks at 111, 396, 512, and 637 cm⁻¹ found in both TiO₂–gC composites and TiO₂ (Fig. S6†) are associated with the TiO₂.^11,28,29 Note that the Raman shifts for carbon species could not be observed in the pristine TiO₂ sample (Fig. S6†), suggesting the absence of carbon species in the pristine TiO₂. Fig. 3d–f show the XPS spectra of Ti2p (Fig. 3d), O1s (Fig. 3e) and C1s (Fig. 3f) as measured on TiO₂–gC (before and after sintering) and TiO₂. Fig. 3d displays Ti2p_3/2 and 2p_1/2 peaks of the TiO₂–gC composite, that appeared at 458.7 and 464.4 eV, respectively. All of the Ti2p peaks of TiO₂–gC composite are found to be negatively shifted to a higher binding energy in comparison with the pristine TiO₂, suggesting a weak interaction between TiO₂ as a core and the graphitic-like carbon as a shell.³⁰ Fig. 3e displays the O1s spectra of TiO₂–gC and TiO₂, where the main peak at 529.9 eV is attributed to the Ti–O bond of the TiO₂ crystal structure. As shown in Fig. 3f, the C1s spectra of the TiO₂–gC composite and controls contain main peaks at around 284.3 and 284.7 eV, corresponding to the C=C bond of sp²-hybridized graphitic carbon and carbonate bonds.³¹ Notably, the intensity area of the C=C and C–C bonds in TiO₂–gC after sintering became prominent compared to the other samples (TiO₂–gC (before sintering), and TiO₂), indicating the formation of a graphitic carbon shell on the surface of TiO₂.³² This XPS result can support the limited particle growth and hindered phase transformation of nanocrystalline TiO₂, owing to bond formation between TiO₂ and carbon. Thus, the TiO₂–gC composite has a unique structure of a porous graphitic carbon framework containing well-dispersed TiO₂ nanoparticles with limited crystalline growth.

Electrochemical performance

Fig. 4 compares the electrochemical performance of the TiO₂–gC composite and controls (TiO₂ and graphitic carbon). All capacities are calculated based on the composite mass. Fig. 4a shows the lithiation/delithiation voltage profiles for the TiO₂–gC composite and controls at a current rate of 0.1 C (1 C = 150 mA g⁻¹) within a cut-off voltage range of 0.0–2.5 V. The initial Li-ion insertion capacities of TiO₂–gC, TiO₂ and gC electrodes are 598, 178 and 263 mA h g⁻¹ at a rate of 0.1 C, respectively. The lithiation/delithiation voltage profiles of TiO₂–gC shows convoluted curves of TiO₂ and gC, demonstrating a high lithiation (598 mA h g⁻¹) and delithiation capacity (469 mA h g⁻¹) in the first cycle. The voltage profile of TiO₂–gC displays a potential plateau at around 1.78 V, a sloped region of 1.78–1.0 V, and a long lithiation tail under 1.0 V in the first lithiation process.¹⁸ The voltage plateaus at 1.78 V for TiO₂ and TiO₂–gC correspond to the phase transition of anatase TiO₂ between the tetragonal and orthorhombic phases during the lithiation reaction, where Li-ions are undergoing insertion/deinsertion within the interstitial sites of TiO₂.^18,23,33 The lithiation capacity originates from the reduction of Ti⁴⁺ to Ti³⁺ with the intercalation of lithium ions, whereas the delithiation capacity comes from the oxidation of titanium.^18,33 The Li insertion/extraction in TiO₂ can be expressed as TiO₂ + xLi⁺ + xe⁻ ↔ Li_xTiO₂, where x is the amount of inserted Li⁺ in TiO₂.¹⁸ In addition, long potential plateaus are found around lower than 0.5 V in the voltage profiles of the TiO₂–gC composite and gC, owing to the intercalation of lithium ions into the carbon layer to form the Li_xC₆ alloy, indicating the formation of graphitic carbon.^18,33 As shown in Fig. S7,† cyclic voltammetry (CV) measurements were also performed for the TiO₂–gC composite, exhibiting consistent results with those of the voltage profiles (Fig. 4a). There was almost no peak position shift observed on the cathodic (−) and anodic (+) scans after the 1^st cycle. Such maintenance of the initial shape of the CV profile suggests a stable lithiation/delithation process.⁵ Fig. 4b and c compare the cycle stability and coulombic efficiency of TiO₂–gC and controls during the cycling, respectively. The cycling test was carried out at 0.1 C for the first two cycles and at 0.5 C for subsequent cycles. The coulombic efficiencies of TiO₂, gC and the TiO₂–gC composite at the first cycle are 55.7, 80.1 and 78.4%, respectively, indicating the larger irreversible capacity of TiO₂–gC and gC compared to TiO₂. The large capacity loss in the first cycle can be mainly attributed to the interfacial reaction between TiO₂ and the electrolyte, which is common to most lithium intercalation hosts. Despite the capacity decay after the first cycle, the TiO₂–gC composite maintains a high lithiation capacity of 589 mA h g⁻¹ in the second cycle and shows quite stable capacity retention in subsequent cycles. In addition, the TiO₂–gC composite has stable capacity retention (>300 mA h g⁻¹) at 0.5 C after 50 cycles and high coulombic efficiency (>95%) during the cycling process, indicating reversible lithium insertion/extraction kinetics. Although there was a capacity drop of TiO₂–gC in the first few cycles, the lithiation capacity still remains at 369 mA h g⁻¹ after 50 cycles, more than twice the value of that of graphitic carbon. Moreover, the TiO₂–gC composite exhibits high coulombic efficiency approaching 98.7% from 30 to 50 cycles. After the second cycle, the coulombic efficiency increases to >98%. These results clearly demonstrate that the TiO₂–gC has a highly reversible capacity and excellent cycle stability during the 50 cycles. It is noteworthy that TiO₂–gC shows a much higher performance than that of previously reported TiO₂ nanostructures, suggesting the superior Li-ion insertion/extraction performance of our novel TiO₂–gC composite. We believe that the enhanced cell performance of the TiO₂–gC composite can be attributed to the synergic effect of TiO₂ and carbon since an appropriate amount of TiO₂ nanoparticles (ca. 60.2%) in the graphitic carbon framework are known to be significantly improving the Li-ion insertion capacity.^11,34 Fig. 4d, e and S8† compare the electrochemical impedance spectroscopy (EIS) of TiO₂, gC and TiO₂–gC after the 1^st cycle and after cycling, respectively. The EIS curves show a semicircle in the high and intermediate frequency ranges and a straight slope line in the low frequency region, indicating the charge transfer resistance at the active material interface (semicircle) and the diffusion of Li⁺ ions in the electrode (sloped line), respectively.^11,34 The relatively smaller semicircle size of TiO₂–gC implies that the charge transfer resistance of TiO₂–gC is also smaller than that of the control electrodes.^8,11 Fig. S8† further confirms no significant EIS change of TiO₂–gC at the 1^st, 50^th and 100^th cycles, indicating the high conductivity of the composite with good structural stability during cycling. We also examined the morphology of the TiO₂–gC composite particles after the cycling test by TEM (Fig. 4f). This image shows retained the morphology of TiO₂–gC after long cycles, implying the robust and stable structure of TiO₂–gC during the lithiation/delithiation process.

Fig. 4 Electrochemical performance of TiO₂, gC and TiO₂–gC: (a) initial voltage profile at 0.1 C, (b) cycle stability test (0.1 C for two cycles and 0.5 C for the subsequent 50 cycles), (c) coulombic efficiency, (d) initial EIS spectra after 1^st cycle, (e) EIS spectra after 50 cycles, (f) TEM image of TiO₂–gC obtained after the cycle test.

Fig. 5 and S9,† furthermore, corroborate the stability of TiO₂–gC for Li-ion insertion by comparing the reversible capacities of TiO₂–gC and the control electrodes (TiO₂ nanocrystal and graphitic carbon) at various current densities (0.1–5 C).

Fig. 5 Electrochemical chemical performance of TiO₂, gC and TiO₂–gC: (a) comparative rate capability test at various current densities (0.1 C–5 C), (b) long term cycle at high C-rate (2 C) test of TiO₂–gC.

Fig. 5a shows the comparative rate capabilities of TiO₂, gC and TiO₂–gC at various current densities ranging from 0.015 to 0.75 A g⁻¹ (0.1–5 C). As expected, although the capacity decreases with increasing current density, the rate capability test shows that TiO₂–gC delivers a high reversible lithiation capacity of 605 mA h g⁻¹ at 0.1 C, 366 mA h g⁻¹ at 0.5 C, and 301 mA h g⁻¹ at 1 C. Even at the high rates of 2 C and 5 C, a high reversible capacity of TiO₂–gC is still achieved, at steady values of 239 and 172 mA h g⁻¹, respectively. The capacity is again recovered to 348 mA h g⁻¹ when the current rate returns to 0.5 C after 50 cycles, indicating good electronic transport throughout the composite. However, the capacity of TiO₂ or graphitic carbon decreases significantly from 126.3 and 272.8 to 15.5 and 109 mA h g⁻¹ under increasing current density (from 0.1 C to 5 C), respectively. In addition, as shown in Fig. S9,† the TiO₂–gC shows superior rate capability to that of TiO₂–C at all current densities (0.1–5 C), implying that the conventional TiO₂–carbon composite is not sufficient to attain good rate capability. The excellent rate capability of TiO₂–gC is mainly attributed to the unique structure of the well-dispersed TiO₂ nanocrystals embedded into the graphitic carbon frameworks, which can facilitate Li-ion charge transport and electron transport, as well as strengthening the carbon–TiO₂ co-constructed channels for lithiated reactions. The superior rate capability of TiO₂–gC compared to TiO₂–C suggests that the conductive graphitized carbon in TiO₂–gC plays an important role in enhancing the conductivity of the composite, thereby contributing to the improved rate capability. The long term cycle stability of the TiO₂–gC composite (Fig. 5b) also supports the excellent electrochemical performance of our material. Remarkably, TiO₂–gC exhibits excellent capacity retention, even when cycled at 2 C 200 times. The TiO₂–gC composites are able to deliver a reversible Li storage capacity of 667 mA h g⁻¹ at the first cycle at 0.1 C, and 252 mA h g⁻¹ can still be retained after 200 cycles at 2 C.

The excellent electrochemical performance (rate capability and cycling stability) of TiO₂–gC compared to the control samples could be achieved through the synergistic effects of the well-developed mesoporosity with a high surface area, the conductive graphitic carbon wall, and the uniformly dispersed TiO₂ nanoparticles, resulting in improved Li⁺ penetration, fast electron transport and high structural stability during cycling. Detailed reasoning is summarized below:

(1) The diffusion length of Li-ions in the TiO₂–gC composite is shortened because of the uniformly dispersed TiO₂ nanoparticles encapsulated in the well-defined porous graphitic carbon framework.^8,11

(2) The electron transport of the TiO₂–gC composite is significantly enhanced due to the conductive graphitic carbon matrix.

(3) The Li⁺ insertion/extraction processes are effectively buffered due to the controlled structural strain/volume change associated with the well-maintained porous structure of the TiO₂–gC composite.

Conclusions

In summary, we demonstrated a facile and scalable method for the fabrication of a mesostructured TiO₂–gC composite via a hydrothermal process, to be employed as a LIB anode material with high performance (high reversible capacity, good cycling stability, and excellent rate capability). The as-formed mesostructured TiO₂–gC composite was composed of TiO₂ nanoparticles uniformly dispersed in a porous graphitic carbon matrix with a high surface area (345.4 m² g⁻¹).

In electrochemical tests, the TiO₂–gC composite exhibited excellent cyclic retention performance of 369 mA h g⁻¹ at 0.5 C after 50 cycles and 252 mA h g⁻¹ at 2 C after 200 cycles, while pristine TiO₂ nanocrystals and graphitic carbon exhibited 35 mA h g⁻¹ and 170 mA h g⁻¹ at 0.5 C, respectively, after 50 cycles. In addition, TiO₂–gC shows excellent rate capability at various current densities of 0.1–5 C. The superior electrochemical performance of the TiO₂–gC composite compared to the controls (gC and TiO₂) could be ascribed to the combined functionalities of TiO₂ and the graphitic carbon as follows; (1) the shortened diffusion length of Li-ions from an interpenetrating structure of a porous carbon framework and well-dispersed TiO₂ nanocrystals, (2) the enhanced electronic conductivity by graphitic carbon and (3) the accommodated volume/strain changes during lithiation/delithiation by the structural stability of TiO₂–gC.

We believe our current approach not only provides a novel way to design and synthesise a high performance TiO₂ composite, but also opens an avenue to prepare a multitude of novel metal oxide based anode materials for future lithium-ion batteries.

Acknowledgements

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (No. 2015R1A2A2A03005789). In addition, this research was supported by C1 Gas Refinery Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Science, ICT & Future Planning (No. 2015M3D3A1A01064957).

Notes and references

1. J. M. Tarascon and M. Armand, Issues and challenges facing rechargeable lithium batteries, Nature, 2001, 414, 359–367.
2. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Tin-based amorphous oxide: a high-capacity lithium-ion-storage material, Science, 1997, 276, 1395–1397.
3. P. G. Bruce, B. Scrosati and J.-M. Tarascon, Nanomaterials for rechargeable lithium batteries, Angew. Chem., Int. Ed., 2008, 47, 2930–2946.
4. A. S. Aricò, P. Bruce, B. Scrosati, J.-M. Tarascon and W. V. Schalkwijk, Nanostructured materials for advanced energy conversion and storage devices, Nat. Mater., 2005, 4, 366.
5. P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J. M. Tarascon, Nano-sized transition-metal oxides as negative-electrode materials for lithium-ion batteries, Nature, 2000, 407(6803), 496–499; H. Sohn, Z. Chen, Y. S. Jung, Q. Xiao, M. Cai, H. Wang and Y. Lu, Robust lithium-ion anodes based on nanocomposites of iron oxide–carbon–silicate, J. Mater. Chem. A, 2013, 1, 4539–4545.
6. S. G. Lee, H. G. Deng, J. Hu, L. H. Zhou and H. L. Liu, Anodic performance of mesoporous TiO₂–C composites with Cu nanoparticles embedded for lithium ion battery, Int. J. Electrochem. Sci., 2013, 8, 2204–2219.
7. P. Y. Chang, C. H. Huang and R. Doong, Ordered mesoporous carbon–TiO₂ materials for improved electrochemical performance of lithium ion battery, Carbon, 2012, 50, 4259–4268.
8. K. Y. Kang, Y. G. Lee, S. Kim, S. R. Seo, J. C. Kim and K. M. Kim, Electrochemical properties of carbon-coated TiO₂ nanotubes as a lithium battery anode material, Mater. Chem. Phys., 2012, 137, 169–176.
9. S. J. Park, H. Kim, Y. J. Kim and H. Lee, Preparation of carbon-coated TiO₂ nanostructures for lithium-ion batteries, Electrochim. Acta, 2011, 56, 5355–5362.
10. S. J. Park, Y. J. Kim and H. Lee, Synthesis of carbon-coated TiO₂ nanotubes for high-power lithium-ion batteries, J. Power Sources, 2011, 196, 5133–5137.
11. L. J. Fu, H. Liu, H. P. Zhang, C. Li, T. Zhang, Y. P. Wu and H. Q. Wu, Novel TiO₂/C nanocomposites for anode materials of lithium ion batteries, J. Power Sources, 2006, 159, 219–222; M.-C. Kim, Y.-W. Lee, S.-J. Kim, B.-M. Hwang, H.-C. Park, E.-T. Hwang, G. Cao and K.-W. Park, Improved Lithium Ion Behavior Properties of TiO₂@Graphitic-like Carbon, Electrochim. Acta, 2014, 147, 241–249.
12. Q. J. Li, J. W. Zhang, B. B. Liu, M. Li, R. Liu, X. L. Li, H. L. Ma, S. D. Yu, L. Wang, Y. G. Zou, Z. P. Li, B. Zou, T. Cui and G. T. Zou, Synthesis of high-density nanocavities inside TiO₂–B nanoribbons and their enhanced electrochemical lithium storage properties, Inorg. Chem., 2008, 47, 9870–9873.
13. H. Zhang, G. R. Li, L. P. An, T. Y. Yan, X. P. Gao and H. Y. Zhu, Electrochemical lithium storage of titanate and titania nanotubes and nanorods, J. Phys. Chem., 2007, 111, 6143–6148.
14. X. P. Gao and H. X. Yang, Multi-electron reaction materials for high energy density batteries, Energy Environ. Sci., 2010, 3, 174–189.
15. M. J. Lindsay, M. G. Blackford, D. J. Attard, V. Luca, M. Skyllas-Kazacos and C. S. Griffith, Anodic titania films as anode materials for lithium ion batteries, Electrochim. Acta, 2007, 52, 6401–6411.
16. G. Sudant, E. Baudrin, D. Larcher and J. M. Tarascon, Electrochemical lithium reactivity with nanotextured anatase-type TiO₂, J. Mater. Chem., 2005, 15, 1263–1269.
17. J.-Y. Shin, D. Samuelis and J. Maier, Sustained lithium-storage performance of hierarchical, nanoporous anatase TiO₂ at high rates: emphasis on interfacial storage phenomena, Adv. Funct. Mater., 2011, 21, 3464–3472.
18. M. Wagemaker, W. J. H. Borghols and F. M. Mulder, Large impact of particle size on insertion reactions: a case for anatase Li_xTiO₂, J. Am. Chem. Soc., 2007, 129, 4323–4327; X. Su, Q. L. Wu, X. Zhan, J. Wu, S. Wei and Z. Guo, Advanced titania nanostructures and composites for lithium ion battery, J. Mater. Sci., 2012, 47, 2519–2534.
19. Y. Yu, L. Gu, C. B. Zhu, P. A. Van Aken and J. Maier, Tin nanoparticles encapsulated in porous multichannel carbon microtubes: preparation by single-nozzle electrospinning and application as anode material for high-performance Li-based batteries, J. Am. Chem. Soc., 2009, 131, 15984–15985.
20. J. Z. Chen, L. Yang, S. H. Fang and S. Hirano, Ordered mesoporous Sn–C composite as an anode material for lithium ion batteries, Electrochem. Commun., 2011, 13, 848–851.
21. J. Hassoun, K. S. Lee, Y. K. Sun and B. Scrosati, An advanced lithium ion battery based on high performance electrode materials, J. Am. Chem. Soc., 2011, 133, 3139–3143.
22. Y. G. Wang, B. Li, C. L. Zhang, T. Hong, S. F. Kang, S. Jiang and X. Li, Simple synthesis of metallic Sn nanocrystals embedded in graphitic ordered mesoporous carbon walls as superior anode, J. Power Sources, 2012, 219, 89–93.
23. W. Wang, Q. Sa, J. Chen, Y. Wang, H. Jung and Y. Yin, Porous TiO₂/C nanocomposite shells as a high-performance anode, ACS Appl. Mater. Interfaces, 2013, 5, 6478–6483.
24. H. S. Bengaard, J. K. Nørskov, J. Sehested, B. S. Clausen, L. P. Nielsen, A. M. Molenbroek and J. R. Rostrup-Nielsen, Steam reforming and graphite formation on Ni catalysts, J. Catal., 2002, 209, 365–384.
25. Y. Wang, Q. Zhu and H. Zhang, Fabrication of β-Ni(OH)₂ and NiO hollow spheres by a facile template-free process, Chem. Commun., 2005, 41, 5231–5233.
26. D. A. H. Hanaor and C. C. Sorrell, Review of the anatase to rutile phase transformation, J. Mater. Sci., 2011, 46, 855–874.
27. L. Liu, Q. Fan, C. Sun, X. Gu, H. Li, F. Gao, Y. Chen and L. Dong, Synthesis of sandwich-like TiO₂@C composite hollow spheres with high rate capability and stability for lithium-ion batteries, J. Power Sources, 2013, 221, 141–148.
28. Y. Ren, Z. Liu, F. Pourpoint, A. R. Armstrong, C. P. Grey and P. G. Bruce, Nanoparticulate TiO₂(B): an anode for lithium-ion batteries, Angew. Chem., Int. Ed., 2012, 51, 2164–2167.
29. F. D. Hardcastle, Raman spectroscopy of titania (TiO₂) nanotubular water-splitting catalysts, J. Arkansas Acad. Sci., 2011, 65, 43–48.
30. G. An, W. Ma, Z. Sun, Z. Liu, B. Han, S. Miao, Z. Miao and K. Ding, Preparation of titania/carbon nanotube composites using supercritical ethanol and their photocatalytic activity for phenol degradation under visible light irradiation, Carbon, 2007, 45, 1795–1801.
31. L.-W. Zhang, G.-B. Fu and Y.-F. Zhu, Efficient TiO₂ photocatalysts from surface hybridization of TiO₂ particles with graphite-like carbon, Adv. Funct. Mater., 2008, 18, 2180–2189.
32. T. Hyeon, S. Han, Y.-E. Sung, K.-W. Park and Y.-W. Kim, High-performance direct methanol fuel cell electrodes using solid-phase-synthesized carbon nanocoils, Angew. Chem., Int. Ed., 2003, 42, 4352–4492.
33. I. Moriguchi, R. Hidaka, H. Yamada, T. Kudo, H. Murakami and N. Nakashima, A mesoporous nanocomposite of TiO₂ and carbon nanotubes as a high-rate Li intercalation electrode material, Adv. Mater., 2006, 18, 69–73.
34. H. Huang, W. K. Zhang, X. P. Gan, C. Wang and L. Zhang, Electrochemical investigation of TiO₂/carbon nanotubes nanocomposite as anode materials for lithium-ion batteries, Mater. Lett., 2007, 61, 296–299.

---

Footnote

† Electronic supplementary information (ESI) available: Supplementary information (Fig. S1–S9) is available. See DOI: 10.1039/c6ra01614f

---