TiO₂ Based Nanomaterials and Their Application as Anode for Rechargeable Lithium-Ion Batteries

Abstract

Titanium dioxide- (TiO2-) based nanomaterials have been widely adopted as active materials for photocatalysis, sensors, solar cells, and for energy storage and conversion devices, especially rechargeable lithium-ion batteries (LIBs), due to their excellent structural and cycling stability, high discharge voltage plateau (more than 1.7 V versus Li+/Li), high safety, environmental friendliness, and low cost. However, due to their relatively low theoretical capacity and electrical conductivity, their use in practical applications, i.e. anode materials for LIBs, is limited. Several strategies have been developed to improve the conductivity, the capacity, the cycling stability, and the rate capability of TiO2-based materials such as designing different nanostructures (1D, 2D, and 3D), Coating or combining TiO2 with carbonaceous materials, and selective doping with mono and heteroatoms. This chapter is devoted to the development of a simple and cost-efficient strategies for the preparation of TiO2 nanoparticles as anode material for lithium ion batteries (LIBs). These strategies consist of using the Sol–Gel method, with a sodium alginate biopolymer as a templating agent and studying the influence of calcination temperature and phosphorus doping on the structural, the morphological and the textural properties of TiO2 material. Moreover, the synthetized materials were tested electrochemically as anode material for lithium ion battery. TiO2 electrodes calcined at 300°C and 450°C have delivered a reversible capacity of 266 mAh g−1, 275 mAh g−1 with coulombic efficiencies of 70%, 75% during the first cycle under C/10 current rate, respectively. Besides, the phosphorus doped TiO2 electrodes were presented excellent lithium storage properties compared to the non-doped electrodes which can be attributed to the beneficial role of phosphorus doping to inhibit the growth of TiO2 nanoparticles during the synthesis process and provide a high electronic conductivity.

1. Introduction

In recent years, lithium-ion batteries (LIBs) have been established as efficient electrochemical energy storage devices and have become the best choice for electric vehicles (EVs) and mobile phones due to their long cycle life, low self-discharge rate, high working voltage, high power and energy density [1, 2]. Developing and using LIBs can significantly reduce pollution of combustion gas by replacing traditional transportation powered by gasoline with environmentally friendly electric vehicles. Following their success in the transport sector, batteries have recently been considered for grid applications, contributing this to the improvement of the energy efficiency of solar, wind, tidal and other clean energy technologies. LIBs are therefore considered to be an essential element in the building an energy-sustainable economy [3, 4].

Figure 1 present the working principle of LIB; both anodes and cathodes could possess a host structure for Li⁺ ions to ensure a good insertion/ disinsertion of these ions during the charge and discharge. The electrolyte is the polypropylene/polyethylene which contains lithium salts (i.e., LiPF₆) in alkyl organic carbonates. The separator, usually Celgard or Whatman, must allow the diffusion of Li⁺ ions between the cathode and the anode during the charging and discharging process [5].

Figure 1.

Working principle of current rechargeable lithium-ion batteries.

The development of large and efficient batteries operating at high potentials necessitates the use of elements that give low an anode intercalation potential. Today, Li⁺ is considered to give the best performances and is therefore widely used. In addition to improving the electrochemical characteristics of anodes, researchers are also concerned with the cost and the environmental impact of the materials under development. In general, an ideal anode material must possess the following characteristics [6, 7]:

• High specific surface area offering more lithium insertion channels,

• Good cycling stability and low volume change during Li ion insertion/desertion process.

• Large pore size for fast Li⁺ ion diffusion and good rate capability,

• Low internal resistance which allows fast charging and discharging process,

• low lithium ions intercalation potential,

• low price and environmental friendliness.

Most commercial LIBs use transition metals oxides or phosphates such as LiCoO₂, LiFePO₄ and LiMnPO₄, as active materials for the cathode, while, the anode is typically made of graphite. Despite its wide commercial use, graphite suffers from a large volume variation during the charge/ discharge process, a low specific capacity, besides safety concerns. To overcome these concerns, TiO₂ is a promising alternative, as it possess excellent structural and cycling stability, high discharge voltage plateau (more than 1.7 V versus Li⁺/Li), high safety, is environmentally friendly, and has a low cost [7, 8]. However, some of the limiting features of this material, including its low electrical conductivity, low capacity and poor rate capability need to be overcome. Figure 2 shows the potential versus Li/Li⁺ and the corresponding capacity density of some potential active anode materials and Table 1 presents a brief comparison between TiO₂ and other anode active materials.

Figure 2.

Potential versus Li/Li⁺ and the corresponding specific capacity of some potential active anode materials for lithium ion batteries.

Materials	Theoretical capacity (mAh/g)	Advantages	Drawbacks
Si	4200	High capacity	Poor cycling, large irreversible capacity
Metal oxides	500–1200	Low cost, High capacity	Low electrical conductivity, low capacity retention and coulombic efficiency
Sn	990	Low cost, good electrical conductivity, good safety	Poor cycling
Graphite	372	Low cost, good working potential	High irreversible capacity, low coulombic efficiency
TiO2	330	Low cost, environmentally friendly, good safety	Low capacity, poor rate capability, low electrical conductivity

Table 1.

Brief comparison between TiO₂ and other anode active materials [5, 9].

Several strategies have been developed to improve the capacity, the cycling stability, and the rate capability of TiO₂-based anodes, and are detailed in the next paragraphs.

1.1 Designing different nanostructured TiO₂

1.1.1 One-dimensional nanostructures (1D)

Nanostructured materials, such as nanotubes, nanowires, nanoneedles, nanofibers and nanorods have been designed for high performance anodes. The interesting performance of 1D TiO₂ was demonstrated by different groups; Tammawat and Meethong studied anatase TiO₂ nanofiber as an anode active material in LIBs, showing a high lithium storage capacity with a stable cycle life and a good rate capability [10]. The excellent performances of these nanostructures could be explained by the increased electronic conductivity, the small nanocrystalline size, the large surface area of the nanofibers and the large Li nonstoichiometric parameters. Another study by Armstrong et al. demonstrated that TiO₂ nanowires exhibit a high capacity of 305 mAh g⁻¹, which is much higher than the capacity value achieved by the bulk TiO₂ (240 mAh g⁻¹) [11]. These improved results are attributed to the large surface area of the prepared nanowires and the good electronic conductivity.

1.1.2 Two-Dimensional Structure (2D)

Compared with Zero-Dimensional (0D) nanoparticles and One-Dimensional (1D) nanostructures, the 2D nanomaterials can store Li⁺ ion in both sides of the structure which offer more exposed surfaces, open charge transport channel for electrolyte penetration and short ion diffusion length [12, 13]. Moreover, the 2D structure is an excellent choice for fast and high lithium storage.

To fabricate 2D TiO₂ materials, significant efforts have been made by several researchers. Li et al. have used hydrothermal methods to synthetize mesoporous TiO₂ nanoflakes (10–20 nm) and evaluate their performance as anode. The electrochemical tests showed that the prepared nanoflakes had a good cycling life and a high discharge specific capacity of 261 mAh g⁻¹ [14]. Another team demonstrated a simple and green synthesis route of anatase petal-like TiO₂ nanosheets. The obtained TiO₂ materials presented a suitable surface area of 28.4 m² g⁻¹, which was proposed to be the reason behind the high capacity and the good cycling stability [15].

1.1.3 Three-Dimensional Porous Structure (3D)

In recent years, 3D porous structure materials have attracted much attention, due to their high porosity, high specific surface area, and low bulk density [16, 17]. Gerbaldi et al. synthesized highly crystalline, nonordered mesoporous anatase TiO₂ with excellent rate capability and cycling stability after prolonged cycling [18, 19]. Lou et al. demonstrated a significantly improved lithium storage capability of TiO₂ hollow spheres and sub-microboxes, together with a high specific capacity, an excellent rate capability, and a long-term cycling stability [20, 21].

1.2 Coating or combining TiO₂ with carbonaceous materials

To improve the electrochemical performance of TiO₂ materials, carbon coating was used in order to reduce the charge transfer resistance, improve the Li⁺, buffer the large volume changes during lithium insertion/extraction, enhance electron transport and prevent the aggregation of active materials [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23]. Xia et al. have reported a carbon-coated TiO₂ anode material with an enhanced electronic conductivity and a low volume expansion during prolonged cycling [24]. In a different study, chemical vapor deposition was used to synthesize TiO₂/CNTs composites, exhibiting a high specific capacity and a long-term cycling stability [25]. This study demonstrated that the enhanced electrochemical performance of this material is due to the structural stability and the efficient conductive network of the TiO₂ particles offered by CNTs. Etacheri et al. mixed TiO₂ with graphene and the synthetized hybrid materials exhibited a high specific capacity, an improved capacity retention and a good rate capability, in comparison with the physical mixture of TiO₂ and reduced graphene oxide [26].

1.3 Selective doping with mono and heteroatoms

To improve the intrinsic conductivity and form more open channels and active sites for Li⁺ transport, doping with cationic or anionic dopants has been shown to be advantageous [27, 28]. Liu et al. evaluated the performance of Ti³⁺ doped TiO₂ nanotube arrays as anode material for LIBs showing an enhanced lithium ion storage performance with an initial discharge capacity of 101 mAh g⁻¹ at a high current density of 10 A g⁻¹ [29]. Furthermore, Sn-doped TiO₂ nanotubes were synthetized by Kyeremateng and coworkers delivering higher capacity values compared to non-doped TiO₂ nanotubes [30]. Otherwise, TiO₂ materials with improved specific capacities were synthesized, by other researchers, using B and N and doping elements [31, 32].

In the following chapter, simple and cost-efficient strategies for the preparation of TiO₂ nanoparticles as anode material for LIBs are discussed. These strategies consist of using the Sol–Gel method, with a sodium alginate biopolymer as a templating agent, and studying the influence of the calcination temperature and the phosphorus doping on the structural, the morphological, the textural and the electrochemical properties of TiO₂ material.

2. Impact of calcination temperature on TiO₂ as anode for rechargeable Lithium-ion batteries

Our group has reported the synthesis of anatase TiO₂ as an anode material of LIBs by a facile synthesis method using a biopolymer as a templating agent. In order to stress the effect of the calcination temperature on the structural, morphological, textural and the electrochemical performances, two heating temperatures were selected: 300°C and 450°C [33]. Titanium dioxide was prepared by a sol–gel method. Sodium alginate powder (1 g) was dissolved by magnetic stirring in 100 mL of distilled water until a gel was formed. This gel was added dropwise to a 100 mL solution of titanium (IV) isopropoxide (0.32 M) and left under stirring for 3 h at room temperature. The obtained solid was collected by centrifugation, washed with distilled water, dried at 70°C overnight and calcined at 300°C and 450°C.

Concerning the structural, textural and morphological observations, all analysis technics resulted in the formation of a pure anatase TiO₂ with aggregated spherical particles. In fact, Figure 3a shows that the unannealed sample present an amorphous like structure, while the diffraction spectra recorded for TiO₂–300 and TiO₂–450 materials are clearly crystalline. The Raman spectroscopy, Figure 3b, confirmed these findings by the presence of three vibration peaks at 632, 508, and 390 cm⁻¹, attributed to Eg, A1g, and B1g modes, respectively, characteristic of TiO₂ anatase phase [34, 35, 36].

Figure 3.

(a) XRD patterns and (b) Raman spectra of TiO₂material obtained at 300°C (black) and 450°C (red).

For the morphological characterization of TiO₂ particles, Scanning Electron Microscopy (SEM) was used. This is shown in Figure 4, where the shapes of the TiO₂–300 and the TiO₂–450 particles are spherical, with an inhomogeneous particles’ size distribution (nano and submicrometric spherical particles). EDX spectroscopy demonstrated, on the other hand, the uniform distribution of Titanium and oxygen.

Figure 4.

SEM images of TiO₂ materials calcined at (a, b) 300°C and (c, d) 450°C.

BET was used to confirm the effect of the calcination temperature on the average pore sizes, resulting in the respective value of 4.4 and 6.0 nm for TiO₂–300, and TiO₂–450 (Figure 5). Both samples were highly porous, which enhances the surface activity for Li⁺ storage and facilitates the liquid electrolyte penetration [33].

Figure 5.

The pore size distribution curves: (black) TiO₂–300°C, (red) TiO₂–450°C.

In order to evaluate the electrochemical performances of TiO₂–300 and TiO₂–450 electrodes, the charge/discharge tests, cyclic voltammetry, Operando XRD of the TiO₂ electrodes were carried out. The charge/discharge profiles of the two electrodes at a current rate of 0.1C are illustrated in the Figure 6. The existence of the cathodic/anodic plateaus located at ∼ 1.7 V (lithiation process) and 1.9 V (delithiation process) are characteristic of the TiO₂ anatase polymorph; tetragonal anatase TiO₂ for the non lithiated TiO₂ and orthorhombic Li_0.5TiO₂ for the Li-rich phase [37, 38]. Besides, the initial reversible capacity of the two electrodes was 266 and 275 mAh·g⁻¹ for TiO₂–300 and TiO₂–450, respectively. TiO₂–300 and TiO₂–450 electrodes demonstrated a coulombic efficiency (CE) of 70% and 75% in the first cycle and a CE higher than 95% in the other cycles, respectively. From the potential vs. capacity profile, it is clearly observed that increasing the synthesis temperature from 300 to 450°C has no obvious impact on the cycling process since this profile is very similar.

Figure 6.

(a) First charge/discharge profiles of TiO₂electrodes calcined at 450°C (red curve) and at 300°C (black curve) cycled between 3.0 and 1.0 V versus Li/Li⁺ at C/10 current rate, (b) cyclic voltammograms of the firstcycle scanned at 0.02 mV s⁻¹.

Concerning the Cyclic voltammetry tests of the as-prepared electrodes (Figure 6), there are a pair of reduction/oxidation peaks at ∼ 1.7 and 1.9 V for both materials, which could be attributed to the Li-ion intercalation/deintercalation in an anatase TiO₂ lattice (Ti⁴⁺ reduction/oxidation). The cathodic/anodic peaks were in accordance with the galvanostatic discharge/charge profiles.

In order to follow the structural evolution of the anatase TiO₂ during the lithiation process, an operando XRD measurement during the discharge/charge of the TiO₂–300°C electrode was carried out. As shown in Figure 7, the (101) reflection peak characteristic of the anatase phase disappeared during the insertion process, which means that the starting material has been successfully lithiated.

Figure 7.

(a) Operando XRD patterns of TiO₂during the 1st discharge from 3.0 to 1.0 V, (b) 1st discharge/charge galvanostatic data at 0.025C currentrate, (c) the 2ϴ region from 24° to 35° showing the disappearance of the (101) reflection peak.

The capacity retention of the two materials is presented in Figure 8. After 100 cycles, TiO₂–300 and TiO₂–450 electrodes showed an excellent capacity retention of 88% and 85%, respectively. Figure 9 presents the rate capabilities of TiO₂ materials evaluated at different current rates at 1.0–3.0 V voltage range. The electrodes were discharged down to 1.0 V and recharged up to 3.0 V at different constant current density from 0.1 to 20 C (1 C = 336 mA g⁻¹). It is clearly observed that the reversible capacity declined gradually with the increase of the current, but still exceeds 73 mAh g⁻¹ even at a rate of 5. These excellent electrochemical properties can be explained by the nanoparticle’s aspect of TiO₂ prepared by biopolymers gelation method.

Figure 8.

Galvanostatic discharge/charge curves vs. Li/Li⁺of (a) TiO₂–300 and (c) TiO₂–450 cycled at a rate of 0.1 C.Cycling performance and coulombic efficiency of (b) TiO₂–300 and (d) TiO₂–450 electrodes cycled between 3.0 and 1.0 V versus Li/Li⁺ at 0.1 C current rate.

Figure 9.

Galvanostatic charge/discharge profiles at different rates of (a) TiO₂–300 and (c) TiO₂–450, rate capability of (b) TiO₂–300 and (d) TiO₂–450 electrodes at variant current rates from 0.1 C to 20 C (1C = 336 mA g⁻¹).

3. Impact of phosphorus doping on TiO₂ as anode for Lithium-ion batteries

Another study by our group have evaluated also the impact of phosphorus doping on the electrochemical performances of TiO₂ as anode material for lithium ion batteries. The phosphorus doped TiO₂ was synthesized using a simple and eco-friendly synthesis method, in which titanium tetra-isopropoxide was used as a titanium precursor and sodium alginate as a complexing agent. The effects of P-doping on the crystal structure, morphology and lithium insertion mechanism were investigated and compared with the undoped TiO₂. Moreover, the P-TiO₂ was tested electrochemically as anode material.

Concerning the synthesis process, TiO₂ and P-TiO₂ materials were synthetized via a gelation of biopolymers, following the synthesis technic proposed by El Ouardi et al. and using phosphoric acid as the phosphorus precursor. To prepare the working electrode, black carbon, PVDF and an active material were mixed in a 7:2:1 wt. ratio. The active material of the first electrode contained pure TiO₂ while the second contained TiO₂ doped at 2% phosphorus.

To identify the crystal structures of TiO₂ and P-TiO₂, XRD was carried out and the results are shown in Figure 10a. As it can be seen in the diffractograms, the diffraction peaks are centered at 25.3^o, 37.9^o, 48.1^o, 54.7^o, 55.0^o, 62.7^o, 68.9^o, 75.04^o and 83.0^o. These are attributed to the (101), (004), (200), (105), (211), (204), (116), (215) and (312) diffraction planes of anatase TiO₂, respectively, indicating that the crystal phase of TiO₂ remained after phosphorylation treatments [39, 40, 41]. At 2θ = 30.7° there is a small peak (*) for the undoped sample, which can be attributed to the existence of the brookite phase, (121) formed during the synthesis [42]. No diffraction peaks that could be attributed to impurities are found in the XRD patterns of TiO₂ and P-TiO₂, suggesting that the sol–gel method can give highly purified anatase TiO₂ products. For the Raman spectra (Figure 10b), the obtained bands at 198, 400, 518, and 641 cm⁻¹ represent the Raman active modes of anatase TiO₂. These results prove that the prepared nanoparticles have an anatase structure; the non-doped TiO₂ sample contained particles with uniform sizes and homogeneous granular surface, while the P-TiO₂ samples remained unchanged. The energy dispersive X-ray (EDX) spectroscopic data of the P-doped TiO₂ demonstrate the uniform distribution of Ti, O and P with no other impurity elements.

Figure 10.

(a) XRD patterns and (b) Raman spectra of non-doped TiO₂(bleu) and P doped TiO₂ (red) materials obtained at 450°C.

The Brunauer–Emmett Teller (BET) method from N₂ adsorption and desorption isotherms carried out at 77 K (Figure 11) showed that both materials presented typical IV adsorption/desorption isotherms with mesoporous structures. Besides, both materials exhibited very similar BET surface, pore size distribution and mesopore diameter. For the absorbance measurement, UV-V spectroscopy showed that the phosphorus doping extended the wavelength response range of TiO₂ into the visible-light region (Figure 12). Moreover, the band gap of TiO₂ and P-TiO₂ was 2.90 and 2.87 eV, respectively. This result shows the effect of phosphorus doping to reduce the band gap and improve the electrotonic conductivity of TiO₂ [43, 44, 45].

Figure 11.

Nitrogen adsorption–desorption isotherm curves and the pore size distribution curves of non-doped and P doped TiO₂materials obtained at 450°C.

Figure 12.

UV-V spectra of non-doped TiO₂ (bleu) and P doped TiO₂ (red) materials obtained at 450°C.

Concerning the electrochemical tests, the charge/discharge curves (Figure 13) show the presence of two plateaus at 1.9 V and 1.7 V for both materials representing cathodic and anodic peaks of anatase TiO₂ nanoparticles, respectively. This charge/discharge process has shown also a good irreversible capacity which does not exceed 10 mAh/g for both materials. For the polarization, P-TiO₂ has shown improved characteristics compared to non-doped TiO₂. This result could be attributed to the improved electronic conductivity.

Figure 13.

(a) First charge/discharge profiles of TiO₂ (bleu curve) and P-TiO₂ (red curve) electrodes calcined at 450°C cycled between 3.0 and 1.0 V versus Li/Li⁺ at C/10 current rate, (b) cyclic voltammograms of the first cycle scanned at 0.02 mV s⁻¹.

Cyclic voltammetry technique was used to study the insertion/extraction properties of lithium ions from the prepared electrodes in the 1.0 and 3.0 V potential window with a scanning speed of 0.02 mV s⁻¹. As it can be seen in Figure 13, there are two cathodic (reduction peak I < 0) / anodic (oxidation peak I > 0) peaks at 1.7 and 1.9 V, respectively, attributed to the insertion / extraction of lithium ions in the TiO₂ nanoparticles. This result is with agreement with the galvanostatic discharge/charge profiles.

For the cycling stability, Figure 14 prove the excellent coulombic efficiency (about 100%) for both TiO₂ and P-TiO₂ electrodes. Besides, these electrodes showed a capacity retention of 78% after 70 cycles and 83% after 90 cycles, respectively. The reason behind these improved electrochemical properties for P-TiO₂ can be the smaller TiO₂ particle size which permits fast lithium insertion / disinsertion process.

Figure 14.

Galvanostatic discharge/charge curves vs. Li/Li⁺ of (a) P doped TiO₂ and (c) non-doped TiO₂ cycled at a rate of 0.1 C. cycling performance and coulombic efficiency of (b) P doped TiO₂ and (d)non-doped TiO₂ electrodes cycled between 3.0 and 1.0 V versus Li/Li⁺ at 0.1 C current rate.

Figure 15 present the rate capabilities of TiO₂ and P-TiO₂ evaluated at different current rates at 1.0–3.0 V voltage range. The electrodes were discharged down to 1.0 V and recharged up to 3.0 V at different constant current density from 0.1 to 20 C (1 C = 336 mA g⁻¹). It is clearly observed that the reversible capacity declined gradually with the increase of the current for both materials, but it still exceeds 80 mAh g⁻¹ for non-doped TiO₂ and 98 mAh g⁻¹ for P doped TiO₂ even at a rate of 5 C. The capacity at 0.1C rate after 70 cycles was recovered to about 185 mAh g⁻¹ for non-doped TiO₂ and 213 mAh/g for P doped TiO₂ after 90 cycles. Thus, indicating the high stability of the anatase TiO₂ nanoparticles and confirming the better performance of P-TiO₂ compared to TiO₂.

Figure 15.

Rate capability of (a) non doped TiO₂ and (b) P doped TiO₂ electrodes at variant current rates from 0.1 C to 20 C (1C = 336 mA g⁻¹).

4. Conclusion

In summary, this chapter show the huge interest in the development and improvement of TiO₂ as anode for high performance rechargeable lithium ion batteries. Several strategies have been developed to improve the conductivity, the capacity, cycling stability, and rate capability of this material, such as designing different nanostructured (1D, 2D and 3D), Coating or combining TiO₂ with carbonaceous materials, and Selective doping with mono and heteroatoms.

Biopolymer gelation is a simple and economically favorable approach for the preparation of TiO₂ nanoparticles. This method that consists of using the Sol–Gel method with a sodium alginate biopolymer as a templating agent showed enhanced performances in comparison with other synthesis techniques. In fact, the prepared TiO₂ electrodes displayed a high specific capacity above 275 mAh g⁻¹ and excellent cycling stability with over 85% capacity retention after 100 cycles. Besides, combining phosphorus doping with this synthesis strategy demonstrated an important discharge capacity of 200 mAh g⁻¹ after 90 cycles under C/10 current rate and has an excellent rate performance. The improved electrochemical performance can be explained based on the P-TiO₂ particles size and band gap modifications upon doping proved by UV-V measurement.

Finally, from this chapter, we can conclude that the use of TiO₂-based materials as anode for commercial lithium ion batteries requires more efforts to overcome the problems encountered, especially the low electrical conductivity, the low energy density, the poor cycling life and the low efficiency.

Acknowledgments

The authors wish to acknowledge Office Chérifien des Phosphates (OCP S.A.) for financial support.