Titanium Dioxide Anode Materials for Lithium-Ion Batteries

Yang Chen; Xiaoli Cui

doi:10.7536/PC200744

Progress in Chemistry ›› 2021, Vol. 33 ›› Issue (8) : 1249-1269 DOI: 10.7536/PC200744

research-article

Scan to read & Share

Titanium Dioxide Anode Materials for Lithium-Ion Batteries

Yang Chen
, Xiaoli Cui ^*

Author information +

History +

PDF

Abstract

Lithium-ion batteries(LIBs) have been widely used in portable electronic devices and electric vehicles owing to their characteristics of high energy density, safety, and long lifetime. Titanium dioxide(TiO₂) is a promising anode material for LIBs due to the advantages of non-toxicity, low cost, abundant sources, and stable chemical structure. However, intrinsic low electronic conductivity and poor lithium-ion(Li⁺) diffusion have restricted its development for practical applications. In this review, we firstly systematically summarize the lithium storage mechanisms of three common TiO₂ polymorphs, that is, a two-phase solid solution of anatase TiO₂, intrinsic pseudo-capacitance of TiO₂(B), and potential-dependent phase transition of rutile TiO₂. Furthermore, to enhance the electron conduction and Li⁺ diffusion, the research progress of nanostructure dimensional tailoring, intrinsic/extrinsic electronic structure manipulation(elemental doping, Ti³⁺ self-doping, and modification of highly conductive materials), and hetero-phase junction optimization are discussed in detail. Finally, the development trend and application prospect of TiO₂ anode materials for LIBs and beyond-LIBs are proposed.

Contents

1 Introduction

2 Structure and lithium storage mechanism of various TiO₂polymorphs

2.1 Crystal structure of TiO₂polymorphs

2.2 Anatase:two-phase solid solution

2.3 TiO₂(B):pseudo-capacitance

2.4 Rutile:potential-dependent phase transition

3 Modification strategies of TiO₂anode materials

3.1 Nanostructure dimensional tailoring

3.2 Intrinsic/extrinsic electronic structure manipulation

3.3 Phase engineering optimization

4 Lithium-ion full batteries performance

5 Conclusions and outlook

Graphical abstract

Keywords

lithium-ion batteries / anode materials / titanium dioxide / nanostructure / electronic structure / hetero-phase junction

Cite this article

Download citation ▾

Yang Chen, Xiaoli Cui. Titanium Dioxide Anode Materials for Lithium-Ion Batteries. Progress in Chemistry, 2021, 33(8): 1249-1269 DOI:10.7536/PC200744

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

Lithium-ion battery is a highly efficient energy storage device with rapid development in recent years, which has the characteristics of high energy density, long life, safety, stability and environmental friendliness, and has been widely used in mobile electronics, electric vehicles and large-scale energy storage. On October 9, 2019, the Royal Swedish Academy of Sciences awarded the "Nobel Prize in Chemistry" to the John B. Goodenough of the University of Texas at Austin, the M. Stanley Whittingham of the State University of New York and Akira Yoshino of Asahi Kasei Co., Ltd. in recognition of their contributions to the development and promotion of Lithium-ion batteries (LIBs), demonstrating the great role of lithium-ion batteries in promoting the development of advanced technologies and the progress of human society^[1]. The development of new electrode materials is one of the keys to further improve the energy density, power density and long cycle stability of lithium-ion batteries.

Anode materials for lithium-ion batteries can be divided into three types: intercalation/deintercalation, alloying/dealloying and conversion according to the difference of lithium storage mechanism. Alloyed/dealloyed anode materials include silicon, germanium and tin, etc. The theoretical capacity of these materials is high, the working potential is 0.005~0.5 V(vs Li/Li⁺ (unless otherwise specified, the potential in this paper is relative to the standard potential of Li), and the volume change during charge-discharge is large (300% ~ 400%). The main reason for restricting the further development and application of these materials is the poor cycle stability^[2]^[3]. Conversion anode materials mainly include some transition metal oxides (such as Fe₃O₄, Fe₂O₃, NiO, MnO, Co₃O₄) and sulfides (such as MoS₂, FeS, SnS₂).The theoretical capacity of this material is 500~1200 mAh·g^-1, but there is also a large volume change during the charge-discharge process, and the cycle performance is poor^[4⇓~6]. At present, intercalation/deintercalation anode materials are the mainstream of practical application, including various carbon materials (graphite, soft carbon, hard carbon, etc.) And titanium-based oxides (lithium titanate, titanium dioxide). Carbon materials are widely used because of their abundant reserves, good stability and low cost, among which graphite is a mature commercial anode material for lithium-ion batteries^[7]. During the charge and discharge process, the volume expansion/contraction of carbon electrode materials is caused by the intercalation/deintercalation of Li⁺, which easily leads to the collapse of crystal structure and the loss of battery capacity. At the same time, the working potential of carbon electrode is close to that of lithium metal, which is easy to form lithium dendrites on the surface when overcharged, causing internal short circuit of batteries, and has potential safety hazards. The development and application of new graphene-based carbon materials is also one of the current research hotspots^{[8⇓⇓⇓⇓~13]}. The titanium-based oxide anode material stores lithium based on the intercalation/deintercalation of lithium ions (Li⁺) into/from the crystal lattice, and has small volume change (less than 4%) in the charge-discharge process and excellent cycle stability compared with alloying/dealloying and conversion materials; And the lithium storage potential is 1.55 to 1.75 V, thereby effectively avoiding the formation of lithium dendrite and improving the safety^[14⇓~16].

Lithium titanate (Li₄Ti₅O₁₂) is a commercialized anode material with good rate capability, but its theoretical capacity is low (175 mAh·g^-1), which can not meet the needs of high energy density secondary batteries. At the same time, the shortage of lithium resources and the rising price of lithium carbonate also limit its practical application. Titanium dioxide (TiO₂) is a precursor for the preparation of Li₄Ti₅O₁₂, which has the advantages of non-toxicity, low price, large reserves and stable chemical structure. It can be used as a structural auxiliary material to limit the volume expansion of silicon (germanium) alloy negative electrode or as a host material of sulfur positive electrode to inhibit the dissolution of polysulfides^{[17⇓⇓⇓⇓⇓⇓~24]}. At the same time, TiO₂ itself is also a promising anode material^[25,26]. The TiO₂ has a plurality of crystal forms, wherein the three most common crystal forms of anatase, TiO₂(B) and rutile have good electrochemical lithium storage activity, and the lithium storage process can be expressed as a formula (1):

(1)$\text{Ti}{{\text{O}}_{2}}+x\text{L}{{\text{i}}^{+}}+x{{e}^{-}}\to \text{L}{{\text{i}}_{x}}\text{Ti}{{\text{O}}_{2}}$

Wherein X is the lithium intercalation coefficient, the maximum lithium intercalation coefficient of the nano anatase TiO₂ and the TiO₂(B) is X = 1, and the theoretical specific capacity is 335 mAh·g^-1; The maximum lithium intercalation coefficient of the nano-rutile TiO₂ is X = 0.85, and the theoretical specific capacity is 285 mAh·g^-1. TiO₂ is a kind of anode material with high safety and long life, which has great application potential and commercial prospects. However, TiO₂ is a wide band gap semiconductor material (the band gap of anatase TiO₂ and TiO₂(B) is 3.2 eV), which leads to its low electronic conductivity (10^-12~10^-7S·cm^-1). At that same time, the limit pore structure and longer Li⁺ migration path in the crystal structure limit the diffusion of the Li⁺ in the bulk phase, and the diffusion coefficient of the Li⁺ is low (10^-15~10^-9cm²·s^-1),The actual capacity of TiO₂ is much lower than the theoretical capacity due to two factors, which limits the practical application of TiO₂ electrode materials^[27]. Therefore, how to improve the electronic conductivity and Li⁺ diffusion coefficient of the TiO₂ itself is an urgent problem to be solved.

In this paper, the characteristic lithium storage mechanisms of anatase, TiO₂(B) and rutile TiO₂ are summarized, and the research progress in improving electron conduction and Li⁺ diffusion is reviewed from three aspects of nanostructure dimension design, intrinsic/extrinsic electronic structure control and phase structure optimization.Finally, the development prospect of TiO₂ anode materials is prospected.

2 Crystal Structure and Lithium Storage Mechanism of TiO₂;

2.1 Crystal structure

There are mainly eight kinds of TiO₂ crystal structures in nature.Including rutile, anatase, bronze (TiO₂(B), brookite, columbite (TiO₂-Ⅱ), manganobite (TiO₂-H), baddeleyite (TiO₂-Ⅲ) and ramsdellite (TiO₂-R)^[28]. The rutile TiO₂ is the most thermodynamically stable, the anatase TiO₂ is relatively stable, and the TiO₂(B) is a metastable structure. These three types of TiO₂ have high electrochemical lithium storage activity. Table 1 lists the corresponding crystal structure parameters and their lithium storage properties^{[25,26,29⇓⇓ ~32]}^[27,28]. The lithium storage properties of other crystal forms of TiO₂ (such as brookite, TiO₂-Ⅱ, TiO₂-H, TiO₂-R) have also been reported, but the mechanism of lithium storage is still unclear^[33,34]^[35]^[36,37]^[38,39].

2.2 Two-phase solid solution mechanism of lithium storage in anatase TiO₂:;

The anatase TiO₂ has a stable crystal structure, belongs to a tetragonal system I41/amd (141) space group, and is formed by connecting an TiO₆ octahedron by sharing a vertex and four edges, and has open channels in the a and B axis directions. Fig. 1A is a potential curve of the discharge process of the anatase TiO₂ hollow sphere at a rate of 0.2 C. The process of inserting the Li⁺ into the lattice of the anatase TiO₂ is roughly divided into three stages^[40].

Region A: a small amount of Li⁺ is embedded into the lattice of the TiO₂, forming the Li_xTiO₂“ "lithium-poor phase". The lithium intercalation coefficient of the material with the micrometer scale is 0 to 0.06, and the lithium intercalation coefficient of the material with the nanometer scale can reach 0.22^[29].

The B:Li⁺ in the region continues to intercalate into the TiO₂ lattice, and the "lithium-poor phase" of the tetragonal system transforms into the "lithium-rich phase" of the orthorhombic system. The two phases coexist, and the intercalation potential remains unchanged (~ 1.75 V) until all the octahedral voids are filled with Li⁺. The Li⁺ in the lattice will repel each other and prevent the further embedding of other Li⁺. In the cyclic voltammograms (CV) curve of the anatase TiO₂ (fig. 1b), an obvious reduction peak appears at about 1.7 V, corresponding to the lithium intercalation plateau where the two phases coexist, and the peak current at different scan rates is proportional to the square root of the scan rate, indicating that the characteristic lithium storage mechanism of the anatase TiO₂ is a lithium intercalation process based on diffusion control^[41].

Under the action of the electric field, the C:Li⁺ in the region is adsorbed on the surface of the TiO₂ to form a pseudocapacitance to store lithium, and the potential curve becomes an inclined curve^[42]. This process is greatly affected by the particle size, and the smaller the particle size of the TiO₂ is, the higher the pseudocapacitance is. Dunn et al. Define the behavior of pseudocapacitance that only appears at the nanometer scale as "extrinsic pseudocapacitance"^[43].

2.3 Intrinsic pseudocapacitance mechanism of lithium storage in TiO₂(B):;

TiO₂(B) is a thermodynamically metastable compound with monoclinic space group C2/m (12), which is connected by TiO₆ octahedra through a common vertex and a common edge. TiO₂(B) has an open channel parallel to the b-axis, and its unit cell density is 3.64 g·cm^-3, which is smaller than that of anatase TiO₂(3.79 g·cm^-3) and rutile TiO₂(4.13 g·cm^-3)^[28]. The open pore structure and low unit cell density are favorable for the migration of Li⁺ in the TiO₂(B), so the TiO₂(B) is considered to be a crystal form with high lithium storage activity^[44,45].

Graetzel et al. Tested the CV curves of the TiO₂(B) at different scan rates, and found that the peak currents of the 1.4 V and 1.5 V reduction peaks (lithiation process) were proportional to the scan rate (Figure 2), thus proposing that the lithium storage mechanism of the TiO₂(B) was different from that of the anatase TiO₂, that is, the anatase TiO₂ was a diffusion-controlled intercalation process (Figure 1b), while the TiO₂(B) was a Faraday pseudocapacitance process^[46]. Dunn et al defined the lithium storage mechanism of TiO₂(B) as "intrinsic pseudocapacitance"^[43].

The Faraday pseudocapacitance behavior of TiO₂(B is attributed to the insertion/extraction process of Li⁺ in the surface/near-surface lattice. The low cell density and open channel structure ensure the rapid migration of Li⁺ from the surface to the near-surface^[43]. The diffusion process of Li⁺ in the lattice of TiO₂(B is limited by the mutual repulsion between Li⁺ in the confined space, so its actual capacity is determined by the steric hindrance caused by the size effect. Using Density Functional Theory (DFT) calculations, Kaghazchi et al studied the diffusion barriers and diffusion coefficients of Density Functional Theory in TiO₂(B) with different sizes^[47]. The TiO₂(B) after lithiation is represented by Li_xTiO₂(B), where X represents the number of moles of Li⁺ intercalated per mole of TiO₂(B)). When only a small amount of Li⁺ is embedded, the mutual repulsion between the (x≤0.025),Li⁺ can be ignored, and the Li⁺ can diffuse along the b-axis or c-axis. When a large number of Li⁺ are embedded in the (0.5≤x≤0.75),Li⁺, the mutual repulsion between them is strengthened, and the Li⁺ diffuse along the B axis, the energy barrier is 0. 32 ~ 0.47 eV, and the average diffusion coefficient is 3.03×10^-10cm²·s^-1. When X = 0.75 (some literatures reported X = 0.71), the block TiO₂(B) reached the maximum intercalation capacity, and the maximum practical capacity was 251 mAh·g^-1^[27,28]. When more Li⁺ are embedded into the (0.75≤x≤1.0),Li⁺, the mutual repulsion is further strengthened, the average energy barrier of Li⁺ diffusion is 0.83 eV, and the diffusion coefficient is reduced to 1.26×10^-16cm²·s^-1. If the Li⁺ is further embedded into the (1.0≤x≤1.25),Li⁺, the average energy barrier increases to 2. 03 eV, and the diffusion coefficient is only 1.15×10^-36cm²·s^-1,Li⁺, which can not be embedded into the TiO₂(B). Therefore, when X = 1.0, the nano TiO₂(B) reaches the maximum lithium intercalation amount, and the theoretical capacity of the TiO₂(B)) is 335 mAh·g^-1.

2.4 Potential-controlled phase transition of rutile TiO₂:;

Compared with anatase TiO₂ and TiO₂(B), rutile TiO₂ has a more stable structure and belongs to the tetragonal P42/mnm (136) space group. The Ti atom is located at the center of the TiO₆ octahedron, the TiO₆ octahedron is arranged in a chain shape along the c-axis, and adjacent chain segments are connected by sharing a vertex; The TiO₆ octahedra in the ab crystal plane are connected in a co-edge manner and occupy half of the octahedral space, and the remaining octahedral space is filled by the Li⁺^[29].

Rutile TiO₂ has an open channel structure only in the c-axis direction, and the diffusion of Li⁺ in the lattice is seriously anisotropic, and the diffusion coefficient of Li⁺ in the c-axis is about 10^-6cm²·s^-1.The diffusion coefficient of Li⁺ in the ab crystal plane direction is only 10^-15cm²·s^-1, so the migration of Li⁺ in the rutile TiO₂ lattice generally occurs only in the c-axis direction^[28]. The pore channel in the c-axis direction is very narrow, which leads to the fact that the rutile TiO₂ bulk material can only be embedded into the 0.1 mol Li⁺ at room temperature; Even at 120 ℃, the rutile TiO₂ only intercalates into the 0.5 mol Li⁺ in the polymer electrolyte solid-state battery. The embedded Li⁺ will also be strongly repelled by other Li⁺, which restricts the further embedding of Li⁺. Frahm et al. Used X-ray absorption fine structure spectroscopy to study the lithium intercalation process of rutile TiO₂^[48]. The results show that the TiO₂ of Li⁺ intercalation into rutile occurs only in a few nanometers of the surface, and further intercalation of Li⁺ can only be achieved by continuously lowering the potential.

The lithium intercalation process of nano-rutile TiO₂ is quite different from that of bulk materials. Maier et al. Studied the lithium storage capacity and cycle performance of nano-rutile TiO₂ in different potential ranges^[49]. The reversible intercalation/deintercalation of lithium ions into/from the nano-rutile TiO₂ occurs in the potential range of 2.5 ~ 1.4 V. When the potential is reduced to 1.0 V, 0.85 mol of Li⁺ can be intercalated, but the process is irreversible. Pan et al. Observed the structural changes of rutile TiO₂ nanowires during lithiation by in-situ TEM^[50,51]. During the first step of lithiation (2.5 ~ 1.4 V), the size of rutile TiO₂ nanowires increases by 140% in the ab plane and decreases by 97% in the c-axis direction, and the unit cell volume changes significantly (Fig. 3A, 3B). Tarascon et al. Also found the change of unit cell parameters of rutile TiO₂ nanoparticles at the potential platform of 1.4 V by in-situ XRD, with the c-axis shortening by 1% and the a-axis increasing by 5%^[52]. During this lithiation, the TiO₂ transforms from rutile to monoclinic (fig. 3D). After 20 cycles, the nanowire size did not change, indicating that the two-phase transition was reversible in the potential range of 2. 5 ~ 1.4 V. With further lithiation to 1.0 V, bubble-like protrusions appear in local regions of the nanowires (fig. 3C), and a second phase transition occurs, forming a rock-salt phase LiTiO₂, and the phase transition process is irreversible (fig. 3D). Ravnsbaek et al. observed a small amount (~ 1%) of niobite-like TiO₂(TiO₂-Ⅱ) structure at the grain boundary of rock-salt phase LiTiO₂ by transmission electron microscopy^[53]. Therefore, the lithium storage process of nano-rutile TiO₂ is closely related to the potential range. In the potential range of 2. 5 ~ 1.4 V, the rutile phase reversibly transforms into the monoclinic phase, and the c-axis channel space is changed by changing the lattice size, thus realizing the reversible intercalation/deintercalation of Li⁺. In the potential range of 2.5 ~ 1.0 V, the lithium storage process of rutile TiO₂ includes two stages (Fig. 3D). The rutile phase first transforms into a Monoclinic intermediate state; After complete lithiation, the monoclinic intermediate phase undergoes an irreversible phase transition to form a Rock-salt phase. The rock-salt phase has a stable LiTiO₂structure and can store lithium reversibly, realizing the intercalation/deintercalation of 0.5 mol of Li⁺.

Wohlfahrt-Mehrens et al. verified the reversibility of two phase transitions of rutile TiO₂ and the reversibility of lithium storage process of monoclinic phase intermediate state and rock salt phase products by controlling the potential interval of cyclic voltammetry test (Fig. 3e, 3F)^[54]. In the potential range of 1. 2 ~ 3.0 V, an obvious reduction peak appears at 1. 4 V in the first cycle, which corresponds to the transition from rutile phase to monoclinic phase, and the phase transition process is reversible in the subsequent lithium extraction/insertion process. In the potential range of 1.0 ~ 3.0 V, two reduction peaks appear at 1.4 V and 1.1 V in the first cycle, corresponding to the transformation of rutile to monoclinic and monoclinic to halite, respectively. The curve shape of the second circle is obviously different from that of the first circle, but it is highly coincident with the subsequent curves, which indicates that the phase transition process from monoclinic phase to halite phase is irreversible, but the halite phase itself can achieve reversible lithium storage.

3 Modification Strategies of Anode Materials for TiO₂;

The development of TiO₂ anode materials is mainly limited by their low electronic conductivity and ion diffusion coefficient, and the actual capacity is much lower than the theoretical capacity (285~335 mAh·g^-1). On the one hand, the diffusion coefficient of Li⁺ in TiO₂ is low, and the migration of Li⁺ in the lattice of TiO₂ is limited, so it is necessary to optimize the structure and particle size. When the particle size is too large, the diffusion path of the Li⁺ is long and the migration rate is low; If the particle size is too small, there will be a large number of surface atoms, which will lead to the reduction of lattice vacancies in the TiO₂ and shorten the platform capacity. On the other hand, the TiO₂ is a wide band gap semiconductor, its conductivity is low, the electron transport rate is slow, and the Li⁺ in the lattice can not be completely removed, resulting in low coulombic efficiency and capacity loss. In order to improve the lithium storage performance of anatase TiO₂, various ways can be taken, such as shortening the migration path of Li⁺ and improving the diffusion coefficient of Li⁺ through structural design and size optimization; The electronic conductivity of the TiO₂ is improved by means of component regulation, surface modification or composite modification, the polarization is reduced, and the reversible capacity is improved.

3.1 Nanostructure dimension design

The nanostructure material has large specific surface area, can increase that contact area of the electrode/electrolyte and improve the interface exchange current density. Nanostructures can also shorten the diffusion distance of Li⁺, make up for the low conductivity of Li⁺, and improve the diffusion coefficient of Li⁺. TiO₂ nanomaterials can be divided into zero-dimensional (0D), one-dimensional (1D), two-dimensional (2D), three-dimensional (3D), multi-dimensional (multi-D) and micro-nano hierarchical structures according to their dimensions and size differences (Figure 4).

3.1.1 Zero-dimensional nanostructures

The TiO₂ of that zero-dimensional nanostructure include quantum dots, nanocrystal, nanoparticles, nanospheres, and the like, and are generally mesoporous in size. The large specific surface area can increase the electrode/electrolyte contact area, improve the interfacial charge and Li⁺ transport efficiency, and show good electrochemical lithium storage activity^[55,56]^[57⇓~59]^{[60⇓⇓~63]}^[64⇓~66].

Wang et al. Prepared graphitized carbon encapsulated mesoporous anatase TiO₂ hollow spheres (H-TiO₂/GC) using SiO₂ nanospheres as hard templates^[65]. Adde a proper amount of ammonia wat into a SiO₂ nanosphere/ethanol solution, and dropwise adding tetrabutyl titanate (TBOT) to coat a layer of amorphous TiO₂ on that surface of the SiO₂; And then coating a layer of carbon on the surface of the mesoporous TiO₂ hollow sphere, and obtaining the mesoporous TiO₂ hollow sphere encapsulated by the graphitized carbon after annealing treatment and 5% hydrofluoric acid etching. The H-TiO₂/GC has high pore volume (0.31 cm³·g^-1), which improves the wettability of the electrode in the electrolyte; The size of the TiO₂ nanocrystals is about 8 nm, which shortens the diffusion distance of the Li⁺. The carbon matrix improves the electronic conductivity of the material and the structural stability of the hollow spheres. The specific capacity of the H-TiO₂/GC is 137 mAh·g^-1 after 1000 cycles at 1 A·g^-1 current density. Stevenson et al prepared mesoporous TiO₂(B) nanoparticles using poly (ethylene oxide) -poly (propylene oxide) -poly (ethylene oxide) triblock copolymer (P123) as a soft template (structure-directing agent), and studied the effect of particle size on the lithium storage properties of TiO₂(B)^[67]. The mesoporous TiO₂(B) is composed of a large number of nanocrystals with a diameter of 4 nm, and the pore diameter is 10 ~ 20 nm. Compared with the nano-TiO₂(B) with a diameter of 250 nm and the bulk material with a particle size of 10 ~ 20 μm, the mesoporous TiO₂(B) has a larger specific surface area, which can increase the electrode/electrolyte contact area and promote the transport of Li⁺ at the interface.

The template-free method can also be used to prepare TiO₂ hollow spheres. Shen et al. Prepared anatase TiO₂ porous hollow spheres by hydrothermal method using titanium tetrachloride (TiCl₄) and ammonium hexafluorotitanate as precursors^[64]. According to the Ostwald ripening mechanism, the crystallinity inside the TiO₂ microspheres is low during the hydrothermal reaction,Under the action of H⁺ and F^-, it is gradually dissolved to form TiF₆^2-, which migrates to the outside, and then recrystallizes to form TiO₂, and finally TiO₂ porous hollow spheres are obtained. At 100 mA·g^-1 current density, the specific capacity of the TiO₂ hollow sphere is 170 mAh·g^-1 after 150 cycles. Anatase TiO₂ hollow spheres were prepared by ethanol solvothermal method using TBOT as precursor^[66]. As shown in Fig. 5, ammonium bicarbonate decomposes into water when heated, which promotes the hydrolysis of TBOT. Under the solvothermal condition of 250 ℃, the TiO₂ hollow spheres are directly formed after 10 H. It is speculated that the formation process is also based on the Ostwald ripening mechanism. The hollow sphere has a reversible specific capacity of 143 mAh·g^-1 after 100 charge-discharge cycles at 300 mA·g^-1 current density. Table 2 summarizes some TiO₂ hollow spheres prepared by hard template, soft template and template-free methods and their lithium storage properties.

3.1.2 One-dimensional nanostructures

The one-dimensional nanostructured TiO₂ has high specific surface area and large aspect ratio, which increases the electrode/electrolyte contact area and provides a one-dimensional migration path for Li⁺ in the bulk phase. Such structures mainly include nanowires, nanorods, nanobelts, nanotubes, etc^{[81⇓⇓~84]}^[85,86]^[87]^{[88⇓⇓⇓⇓~93]}.

Tarascon et al. studied the electrochemical properties of rutile TiO₂ nanorods^[52]. The initial capacity of commercial nano-rutile TiO₂ (particle size about 50 nm) is 77 mAh·g^-1(0.23Li/TiO₂), and the stable capacity is only 37 mAh·g^-1(0.11Li/TiO₂). The rutile TiO₂ nanorods (length: 200 nm, diameter: 10 nm) synthesized by the liquid phase precipitation method have an initial capacity of 150 mAh·g^-1(0.45Li/TiO₂) after 285 mAh·g^-1(0.85Li/TiO₂),60 charge-discharge cycles, which is attributed to the formation of a highly reactive rock-salt LiTiO₂ in the first cycle. Chen et al. Used a template-free method to directly grow rutile TiO₂ nanorods on the surface of titanium sheets^[94]. The nanorods are oriented along the c-axis and perpendicular to the substrate, with a diameter of 40 ~ 50 nm, a length of 1.5 ~ 2.0 μm and an average loading of 0.5 mg·cm^-2. Compared with the compact surface layer, the oriented nanorods can effectively increase the contact area between the electrode and the electrolyte, accelerate the charge transfer and ion conduction at the interface, and increase the actual specific capacity by 10 times.

Cui et al. Prepared anatase TiO₂ nanoribbons by hydrogen peroxide (H₂O₂) assisted alkali-free wet chemical method using titanyl sulfate (TiOSO₄) as precursor^[87]. A proper amount of TiOSO₄, glycine and nitric acid were added to a crucible containing 10 mL of deionized water and quickly transferred to a furnace preheated to 400 ℃ for 15 min to obtain a black intermediate. After ultrasonic dispersion, the black intermediate was dispersed in the H₂O₂ solution and kept at room temperature for 72 H, and then the centrifuged product was annealed at 400 ℃ in air for 1 H to obtain the TiO₂ nanobelts. The one-dimensional nanoribbon structure can promote that rapid transmission of electron along the radial direction; Meanwhile, the thickness of the nanoribbons is less than 5 nm, which shortens the migration distance of the Li⁺. The TiO₂ nanobelt has a porous structure, and the pore volume is 0.666 cm²·g^-1, so that the electrode is fully soaked by the electrolyte, and the interface transfer rate of the Li⁺ is improved. The high specific surface area (119 m²·g^-1) of nanoribbons can increase the electrode/electrolyte contact area and promote the interfacial charge transfer. In addition, the flexible porous nanoribbons can also alleviate the volume/stress changes during charging and discharging, and improve the cycle stability of the battery.

Compared with other one-dimensional nanostructures, nanotubes can release the local stress caused by the volume change in the process of lithium intercalation and deintercalation, and the tube wall can also provide a larger contact area between the electrolyte and the electrode, so one-dimensional TiO₂ nanotubes usually have better electrochemical performance, which has attracted wide attention of researchers^[95]. Chen et al. Prepared ultra-long TiO₂(B) nanotubes with a length of about 30 μm by a magnetic stirring-assisted hydrothermal method^[91]. Magnetic stirring accelerates the dissolution/recrystallization process of TiO₂(B) nanoparticles, which is helpful for the preparation of ultra-long (TiO₂(B)) nanotubes. The specific surface area of the nanotube is 130 m²·g^-1, which can increase the contact area between the electrode and the electrolyte and improve the migration efficiency of the Li⁺ at the interface; Meanwhile, the nanotube structure can shorten the migration distance of the Li⁺ and improve the diffusion coefficient of the Li⁺. The TiO₂(B) nanotube has excellent rate capability, and the reversible capacity is 114 mAh·g^-1 after being cycled for 10, 000 times at a current density of 8.4 A·g^-1.

3.1.3 Two-dimensional nanostructures

Two-dimensional materials are the current research focus of nanomaterials, and two-dimensional TiO₂ nanostructures mainly include nanosheets, nanodisks and so on^{[96⇓⇓~99]}^[100]. The material can increase the proportion of high-activity crystal faces through controllable growth, provide more lithium intercalation active sites, and improve the lithium storage performance.

Lou et al. Prepared anatase TiO₂ nanosheets with high (001) crystal face ratio by hydrofluoric acid-assisted hydrothermal reaction using titanium tetraisopropoxide (TTIP) as precursor^[96]. The nanosheets have a rectangular or square morphology with a side length of 20 ~ 100 nm and a thickness of 10 nm. The proportion of (001) crystal plane of the TiO₂ is 62%, and the high proportion of (001) crystal plane can provide more active sites for the insertion of Li⁺; The thin layer structure can reduce the steric hindrance and accelerate the rapid desorption of the Li⁺. In addition, the surface of the nanosheet is doped with fluorine (F), and the atomic ratio of F/Ti is 0.21. F doping can reduce the band gap of the TiO₂ and improve the electronic conductivity. Recently, Chou et al. Prepared ultrathin two-dimensional mesoporous TiO₂/rGO composites by layer-by-layer assembly technology, which had a reversible lithium storage capacity of 350 mAh·g^-1 at 200 mA·g^-1 current density and a specific capacity of 245 mAh·g^-1 after 1000 cycles at 1 A·g^-1 current density^[99]. Choi et al. Prepared ultrathin anatase TiO₂ nanodisks with a thickness of 0.5 ~ 1.0 nm by a sol-gel method occurring at the organic solvent/water interface^[100]. The method comprises the following steps of: dissolving TTIP in an n-hexane solvent, adding ice cubes, wherein the TTIP is hydrophobic, and a hydrolysis reaction occurs at an n-hexane/water interface when the TTIP meets a small amount of water on the surface of the ice, and nanoparticles are horizontally aggregated at the interface to form a nanodisk structure; and finally, removing the organic solvent and water by using ethanol, and drying to obtain the ultrathin TiO₂ nanodisk. The reversible capacity of the nanodisk is 112 mAh·g^-1 at 3.4 A·g^-1 current density.

3.1.4 Three-dimensional nanostructures

Three-dimensional TiO₂ nanostructures mainly include mesoporous microspheres, hollow cages, arrays, reconstructions and so on^[69,71,101]^[73,102,103]^[104,105]^{[106⇓⇓~109]}. Ren Hao et al. Summarized the synthesis of hollow multi-shell TiO₂ by hard template, soft template and template-free methods and their applications in energy storage and conversion such as lithium-ion batteries and photocatalysis^[110].

The three-dimensional skeleton can maintain the stability of the structure during the cycle. O'Dwyer et al. Prepared a continuous regular network of rutile TiO₂ with inverse opal structure by using polystyrene as template and titanium tetrachloride tetrahydrofuran complex (O 'Dwyer) as precursor^[111]. The thickness of the skeleton wall of the network structure is 40 nm, and the diameter of the holes is 330 nm. At 450 mA·g^-1 current density, after 100 cycles, the size of the hole wall and the hole is basically the same as the initial position; After 1000 cycles, the pore wall thickness increased to 130 nm, with an average increase of 0. 1 nm per cycle, and the pore diameter decreased to 240 nm; after 5000 cycles, the network structure remained intact, with a pore wall thickness of 160 nm, an average increase of 7. 5 pm per cycle, and the pore diameter decreased to 210 nm. Therefore, the inverse opal structure rutile TiO₂ has excellent structural stability. At 75 mA·g^-1 current density, the capacity of the rutile TiO₂ after 100 cycles is 170 mAh·g^-1,1000, and the capacity after 10 cycles is still 140 mAh·g^-1.

The pore structure in the three-dimensional framework can make the electrolyte fully infiltrate the electrode and accelerate the interface transfer rate of the Li⁺, thereby improving the rate capability. Nan et al prepared TiO₂/SiO₂ composite fibers using tetrabutyl titanate and tetraethyl orthosilicate as precursors, and then subjected to alkali solution etching, ion exchange and annealing to obtain layered porous fibers assembled by anatase TiO₂ nanosheets^[106]. At 5.04 A·g^-1 current density, the specific capacity of the material is 102 mAh·g^-1 after 1000 charge-discharge cycles, and the capacity retention rate is 80%, showing excellent cycle stability. The three-dimensional interconnected porous network structure provides enough space for the TiO₂ nanoribbons to be fully soaked in the electrolyte, thereby improving the migration rate of the Li⁺ at the electrode/electrolyte interface. The ultra-thin nanosheet shortens the diffusion path of the Li⁺ and provides sufficient active sites for the migration of the Li⁺ on the surface. Chen et al. prepared defect-rich TiO₂(B) nanosheets through a rapid dissolution-recrystallization hydrothermal process and assembled them into a 3D skeleton structure. The hierarchical 3D structure and a large number of defects gave the material excellent rate capability and good cycling performance. The specific capacity retention rate of the material after 6000 cycles at 6.7 A·g^-1 current density was 96.1%^[112].

3.1.5 Multidimensional composite nanostructures

Single-dimensional TiO₂structural units construct multi-dimensional composite structures in the form of self-assembly, composition or reconstruction through "bottom-up" or "top-down"^{[113⇓⇓~116]}. Multi-dimensional composite has more active sites for lithium storage than single-dimensional composite, and can increase the specific surface area and improve the transport efficiency of interfacial charge and Li⁺.

The self-assembly process is the most common form of "bottom-up", which can assemble nanowires/bundles/rods into spherical clusters to exert the high specific capacity properties of nanomaterials while maintaining high coulombic efficiency^[117]. Zhang et al. prepared rutile TiO₂ clusters using titanyl sulfate as raw material through hydrothermal reaction under acidic conditions and subsequent heat treatment, and applied them to lithium-ion batteries with coulombic efficiency greater than 99.3%^[118]. They also studied the influence of solvents on the morphology of the samples. Water was the initial solvent, and the products were spherical clusters assembled by nanobeams. Ethylene glycol (EG) was used as the initial solvent, and the spherical clusters were assembled from nanowires. The capacity of rutile TiO₂ nanowire clusters reaches 190 mAh·g^-1 after 200 cycles at 0. 6 C rate. Hu et al. Also prepared rutile TiO₂ clusters by hydrothermal method using titanium trichloride (TiCl₃) as titanium source and a mixed solution of water and ethylene glycol as solvent^[119]. The mixed solvent can adjust the nucleation and growth process of the TiO₂ to prepare the TiO₂ nanowires with uniform size and monodispersity, and the diameter of the nanowires in the cluster can be reduced by increasing the mass ratio of the EG/H₂O. The reversible capacity of the material reaches 140 mAh·g^-1 after 300 charge-discharge cycles at 10 C rate. Similar structures include dandelion, hedgehog and other morphologies assembled by nanowires/bundles/rods with a diameter of about 20 nm^[120⇓~122]. These clusters have high specific surface area, which can increase the contact area between the electrode and the electrolyte, and the nanowire/bundle/rod structure itself can shorten the diffusion distance of the Li⁺.

The second treatment completes the "top-down" reconstruction process, which can expand the one-dimensional nanostructure to a multi-dimensional composite structure, improve the active sites of lithium storage, and accelerate the interfacial charge transfer. Huang et al. Prepared a vertically oriented TiO₂ nanotube array on a titanium substrate by anodization, with an outer diameter of 130 nm and a wall thickness of 20 nm (Fig. 6a)^[113]. After the TiO₂ nanotube array was immersed in water for 72 H, the thickness of the nanotube wall decreased, the surface became rough, and a large number of nanoparticles were generated, resulting in an anatase TiO₂ nanotube/nanoparticle composite array (fig. 6B). The nanotube structure is beneficial to the rapid migration of electrons in the vertical orientation and promotes the interfacial charge transfer; Compared to the TiO₂ nanotube arrays, the TiO₂ nanotube/nanoparticle composite arrays have a higher specific surface area and provide more active sites for the intercalation/deintercalation of Li⁺ (fig. 6C). At 0.02 mA·cm^-2 current density, the capacity of the TiO₂ nanotube/nanoparticle composite array is 0.13 mAh·cm^-2 after 50 charge-discharge cycles, which is 1.3 times that of the TiO₂ nanotube array.

3.1.6 Micro-nano hierarchical structures

Although nanostructures can make up for the low ionic conductivity by shortening the diffusion path of Li⁺, and the size effect of nanostructures can improve the surface pseudocapacitance contribution, the average coulombic efficiency of nanomaterials with high specific surface area is usually reduced. Zhou et al. Found that at 0.05 mA·g^-1 current density, the discharge capacities of commercial rutile TiO₂ powders with particle sizes of 300, 30 and 15 nm after 20 charge-discharge cycles were 50, 137 and 206 mAh·g^-1, respectively, while the coulombic efficiencies were 99.8%, 95.5% and 93.2%, respectively^[123]. The presence of active sites on the surface of nanomaterials can also reduce the first-pass coulombic efficiency, such as Ti-OH and Ti-O active sites on the surface of nanomaterials, which lead to electrolyte decomposition and irreversible Li⁺ capture^[124].

In recent years, secondary granulation has been successfully applied to the modification of electrode materials such as ternary positive electrode, polyanionic positive electrode, silicon negative electrode and oxide negative electrode for lithium ion batteries^{[125⇓⇓⇓⇓~130]}. The secondary granulation is to uniformly mix the primary particles with a proper amount of binder, additives and auxiliaries through mechanical stirring, hydrothermal treatment and other ways under the heating condition, and to obtain micron-sized secondary particles with uniform size and larger particle size in combination with subsequent heat treatment^[126,131]. Secondary granulation can effectively improve the liquid retention performance of the material and the tap density of the electrode, reduce the expansion rate of the material, and reduce the pores between the primary particles and the active sites of surface oxygen-containing groups, so the secondary particles usually have good cycle stability and high first coulombic efficiency^[129⇓~131]. Yoshida et al. Prepared micron-sized spherical particles by secondary granulation, which reduced the number of surface active sites and increased the coulombic efficiency to 90% for the first time^[129]. Secondary granulation is different from the "bottom-up" self-assembly process and "top-down" reconstruction process of nanomaterials. The latter (secondary treatment) obtains multi-dimensional composite materials from single-dimensional nanomaterials. High specific surface area and nano-size effect can improve the mass specific capacity and rate performance of materials. The former (secondary granulation) can obtain larger particle size micro-nano structure materials, which are usually zero-dimensional spherical particles with small specific surface area, high tap density, stable structure and few surface side reaction active sites, and can effectively improve the volume specific capacity, cycle stability and coulombic efficiency of the materials^{[127,128,132]}.

Therefore, the size effect can be optimized and the electrochemical lithium storage activity of the material can be improved by comprehensively using the above technical means to regulate the nanostructure dimension.

3.2 Intrinsic/extrinsic electronic structure control

The dimensional design of TiO₂ nanomaterials can improve the diffusion coefficient of Li⁺ at the interface and in the bulk phase. However, the electronic conductivity of the TiO₂ itself is still low, and the intrinsic or extrinsic methods are usually used to improve its electronic conductivity. The intrinsic mode is mainly through element doping and Ti³⁺ self-doping (oxygen vacancy) to introduce impurity energy levels, improve the intrinsic electronic structure and increase the carrier concentration. The extrinsic way is mainly to improve the overall electronic conductivity of the composite by surface coating, embedding or compounding of highly conductive materials without changing the intrinsic electronic structure of the TiO₂.

3.2.1 Element doping

Element doping is to dope metal/non-metal elements without changing the crystal structure of the TiO₂.Replace the Ti⁴⁺ position (Cu⁺, Mn²⁺,Co²⁺ 、Ni²⁺ 、Zn²⁺ 、Al³⁺ 、Cr³⁺ 、Fe³⁺ 、Sm³⁺ 、USn⁴⁺ UNMn⁴⁺Zr⁴⁺ 、UNb⁵⁺ UNV⁵⁺ NKMo⁵⁺/Mo⁶⁺ K1P⁵⁺S⁶⁺) or UO^2- (UNB^-F^- 、US^2- UNN^3-An electron or hole defect energy level is introduce to improve that conductivity of the material^[61]^[98]^[133]^[134]^[135]^[136]^[137,138]^[139,140]^[141]^[142~144]^[145]^[146]^[147~150]^[145]^[151]^[152]^[140]^[81,153]^[96,154]^[140]^{[138,155~158]}. Element doping includes substitutional doping and interstitial doping, and substitutional doping is the most effective, but the amount of doping is limited by thermodynamic solubility. Interstitial doping generally occurs on the surface of the sample, which has limited effect on improving the carrier mobility.

Chen et al. studied the effects of N doping on the conductivity, crystal structure and electrochemical properties of TiO₂(B) nanowires^[156]. The N^3- preferentially occupies the interstitial position of the TiO₂(B), and the maximum interstitial ratio is 0.55 at.%. The doping of interstitial N^3- has little effect on the crystal structure and conductivity. Furthermore, N^3- can replace 1. 3 at.% of the O^2- sites, strengthen the Ti — O — Ti and O — Ti — O bonds, and improve the stability of the TiO₂(B) crystal structure. The conductivity of the pure TiO₂(B) is 3.0×10^-11S·cm^-1, and the conductivity of the substituted position N^3- doped TiO₂(B)) is 4.6×10^-8S·cm^-1, which is improved by three orders of magnitude. Recently, Bang et al. Found through theoretical calculations that when there are oxygen vacancies in the system, nitrogen doping in interstitial positions is more conducive to improving the electronic conductivity of TiO₂ than nitrogen doping in substitutional positions^[157]. Gao et al., combined with theoretical studies, show that N-doping can reduce the band gap of the TiO₂(B), thereby increasing the electronic conductivity, but the formation energy of the Li_xTiO₂(B) is higher^[159]; F doping has lower formation energy, reduces the diffusion barrier of the Li⁺, and improves the diffusion ability of the Li⁺ in the crystal lattice. Wang et al. Prepared TiO₂(B) nanowire arrays directly on titanium foil by hydrothermal reaction-ion exchange, and then obtained phosphorus-doped (TiO₂(B)) nanowire arrays by heat treatment in sodium dihydrogen hypophosphite atmosphere^[152]. The results of DFT calculation show that some of the Ti⁴⁺ are replaced by the P⁵⁺. The P⁵⁺ doping improves the surface lithium affinity of the TiO₂(B) nanowires, so that the nanowires have higher surface reaction activity. After 3000 charge-discharge cycles at 10 C rate, the capacity of the phosphorus-doped TiO₂(B) nanowire array remains 152.6 mAh·g^-1, and the capacity retention rate is 80% relative to the initial capacity.

Metal/nonmetal co-doping can further improve the conductivity of the TiO₂. Lai et al. Prepared Fe-S co-doped anatase TiO₂ nanorods by sol-gel method using ferric chloride hexahydrate as iron source and thiourea as sulfur source^[140]. The ionic radius of the Fe³⁺ (0.650 650 Å) is slightly larger than that of the Ti⁴⁺ (0.606 606 Å), and the Fe³⁺ takes the place of the Ti⁴⁺. The existence of S — O — S bond was detected by X-ray photoelectron spectroscopy, which indicated that most of the S atoms were in the form of S⁶⁺ replacing Ti⁴⁺, except for a small amount of S^2- replacing O^2-. Dai et al. Prepared Cr-N co-doped anatase TiO₂ mesoporous microspheres by hydrothermal reaction-ammonia annealing method^[138]. Cr³⁺ and N^3- take that place of Ti⁴⁺ and O^2-, respectively,The electronic conductivity of Cr-N co-doped TiO₂ is 1.7 times that of N-doped TiO₂ and 5.1 times that of pure TiO₂. Cr-N co-doping not only improves the electronic structure of the TiO₂, but also increases the diffusion coefficient of the Li⁺. The Li⁺ diffusion coefficient of Cr-N co-doped TiO₂ is 4.0 times of that of N-doped TiO₂.

The actual doping amount of different systems needs to be optimized to achieve high specific capacity and high cycle stability. Ghule et al. Studied the effects of different doping amounts of Cu⁺ on the structure and electrochemical properties of anatase TiO₂^[61]. The substitution of Cu⁺ for Ti⁴⁺ results in the formation of oxygen vacancies, which improves the electron conductivity. When the doping amount is 7%, the electrode/electrolyte interface has the smallest charge transfer resistance and the highest electronic conductivity. At this time, part of Cu will exist in the form of Cu₂O. The Cu₂O reacts with Li to form Cu and Li₂O, and the process is reversible, so the specific capacity of the 7% Cu-doped TiO₂ is also significantly improved. Zheng et al. Prepared Sn-doped anatase TiO₂ thin films by evaporation-induced self-assembly method^[142]. Appropriate amount of Sn doping can improve the electronic conductivity of the TiO₂, but excessive Sn doping will exist in the form of SnO₂ and phase separation with the TiO₂. The SnO₂ reacts with Li to form Sn and the Li₂O,Sn further reacts with Li to form Li_xSn alloy, which affects the cycle stability of the battery. When the Sn content is 6%, the reversible specific capacity is the highest, and the discharge capacity is 252.5 mAh·g^-1 after 80 cycles at 84 mA·g^-1 current density.

In addition to substitution doping, interstitial doping is also a form of element doping. Ye et al. Directly annealed the polyaniline /TiO₂ composite in the air to prepare the C-N co-doped TiO₂^[160]. C-N co-doping is mainly at the interstitial sites of the surface layer (Fig. 7A), rather than the substitution sites of the Ti⁴⁺ or O^2-. The interstitial doping reduces the band gap of the TiO₂, which is beneficial to the interfacial charge transfer. The reversible specific capacity of the C-N co-doped TiO₂ remains 156.7 mAh·g^-1 after 150 charge-discharge cycles at a current density of 1 A·g^-1. Bang et al. Found that in the presence of oxygen vacancies, interstitial N doping is more conducive to improving the electronic conductivity of TiO₂ than substitutional N doping^[157]. By controlling the sequence of reduction/oxidation atmosphere in the annealing process, the internal interstitial N-doped TiO₂ (Fig. 7B) was successfully prepared, and its lithium storage performance was significantly better than that of the surface interstitial N-doped TiO₂.

Therefore, element doping can improve the intrinsic electronic structure of the TiO₂, increase the number of carriers by introducing doping elements, and achieve rapid electronic conduction in bulk and surface.

3.2.2 Ti³⁺ self-doping (oxygen vacancy)

Oxygen vacancy is a kind of defect $\text{(}{{\text{O}}_{O}}\to {{V}_{O}}^{..}+2{{e}^{-}}+\frac{1}{2}{{\text{O}}_{2}})$ formed by the lack of lattice oxygen in oxygen-containing compounds under certain conditions. By binding one electron $({{V}_{O}}{{^{.}}^{.}}+{{e}^{-}}\to {{V}_{O}}^{.})$ or two electrons $({{V}_{O}}^{..}+2{{e}^{-}}\to V_{O}^{X})$, the bound one-electron oxygen vacancy $({{V}_{O}}^{.})$ and the bound two-electron oxygen vacancy $(V_{O}^{X})$ are obtained respectively^[161]. Scanning tunneling Electron microscopy (STM) or electron paramagnetic resonance (EPR) is usually used to identify the existence and state of oxygen vacancies in materials. The g-factors of EPR for the bound singlet oxygen vacancy $({{V}_{O}}^{.})$ are generally in the range of 2.001 to 2.004, while the bound doublet oxygen vacancy () and the unbound oxygen vacancy $({{V}_{O}}^{..})$ can not be directly detected by EPR^[161⇓~163]. Due to the mutual repulsion between free electrons, the bulk of the TiO₂ is usually a ${{V}_{O}}^{.}$, while the surface and near-surface are mostly $V_{O}^{X}$.However, the $V_{O}^{X}$ can reduce the adjacent Ti⁴⁺ to the Ti³⁺ and obtain the ${{\text{V}}_{O}}^{.}(V_{O}^{X}+T{{\text{i}}^{4+}}\to {{\text{V}}_{O}}^{.}+T{{\text{i}}^{3+}})$, and the generated Ti³⁺ can be identified by EPR, and the g-factor is generally between 1.94 and 1.99^[164]^[165⇓~167]. It should be noted that the Ti³⁺ on the surface of the material is unstable and easy to be oxidized by O₂ in the air to generate paramagnetic $O_{2}^{-}(T{{\text{i}}^{3+}}+{{O}_{2}}\to T{{\text{i}}^{4+}}+O_{2}^{-})$, and the corresponding EPR is G = 2.003, so the detection of the Ti³⁺ needs to be carried out under oxygen-free and low-temperature conditions^[168]. In addition, XPS analysis can also be used to characterize the composition of the Ti³⁺ on the surface (< 10 nm) of the material, which can indirectly infer the existence of surface oxygen vacancies. Ti³⁺ self-doping or oxygen vacancies are introduced into the lattice of the TiO₂ by metal reduction, hydrogen reduction, electrochemical reduction and other methods to act as charge transfer carriers and improve the electronic conductivity of the TiO₂^{[169⇓⇓~172]}^{[173⇓⇓~176]}^[177,178]. The undoped TiO₂ is usually a white powder, while the Ti³⁺ self-doped TiO₂ appears in darker colors such as yellow, blue, gray, and black^[174,179]^{[169,180⇓⇓ ~183]}^[184]^{[170,178,185,186]}.

Ji et al. Prepared Ti³⁺ self-doped rutile TiO₂ by solvothermal method using TiCl₃ solution as precursor, magnesium powder as reducing agent and isopropanol as solvent^[170]. Similarly, Lee et al. Also prepared Ti³⁺ self-doped rutile TiO₂ by solvothermal method using zinc powder as a reducing agent^[171]. The magnesium powder or zinc powder reducing agent inhibits part of the Ti³⁺ from being oxidized into Ti⁴⁺ and generates black or blue Ti³⁺ self-doped TiO₂. Xu et al. Prepared blue Ti³⁺ self-doped anatase TiO₂ nanoparticles by aluminothermic reaction using anatase TiO₂ nanosheets as precursors, and the particle size was reduced from 60 nm to 31 nm^[169]. Zhang et al. Prepared TiO₂ nanosheets of different sizes by hydrofluoric acid-assisted hydrothermal method, and prepared hydrogenated TiO₂ powder by hydrogen reduction method to obtain gray TiO₂(H-90), blue TiO₂(H-60) and black TiO₂(H-30), respectively^[176]. H-90, H-60 and H-30 are round or oval particles with diameters of 61, 21 and 15 nm, respectively. The smaller the particle size is, the higher the concentration of Ti³⁺ (oxygen vacancies) is, and the higher the conductivity is. The smaller the particle size is, the shorter the diffusion distance of the Li⁺ is and the higher the diffusion coefficient of the Li⁺ is. The specific capacity of H-30 sample after 100 cycles at 1 C rate is 226 mAh·g^-1, which is 2. 7 times of the reversible capacity of pure TiO₂. The self-doping of the Ti³⁺ improves the electronic conductivity of the TiO₂, and the reduction of the particle size improves the Li⁺ diffusion coefficient,The reversible capacity of the Ti³⁺ self-doped TiO₂ reaches 165 mAh·g^-1 after 100 cycles at 1 C rate.

At present, the main methods for preparing the self-doped TiO₂ of the Ti³⁺ are the above-mentioned metal thermal reduction, hydrogen reduction and metal chemical reduction, and other methods include solid phase reaction, electrochemical reduction, amino chemical reduction, laser irradiation and ultrasonic treatment^[182]^[177,178]^[179]^[187]^[188]. Chiu et al used calcium titanate and TiCl₄ as precursors to prepare TiO₂/CaCl₂ composite intermediates by evaporation-solid state reaction (700 ℃), and then washed with water to obtain blue Ti³⁺ self-doped rutile TiO₂ nanowires^[182]. Recently, by using the reducibility of amino functional groups in biomass templates, Liu et al. Prepared hierarchical porous oxygen-deficient TiO₂ through a simple pyrolysis method, and the samples changed from weak crystalline phase to amorphous phase during lithiation.With the color changing from yellow to black, it is speculated that the redox reaction between Ti³⁺ and Ti occurs, and the specific capacity still reaches 206 mAh·g^-1 after 8000 cycles at 10 A·g^-1 current density^[179]. Hydrogen reduction, metal thermal reduction and electrochemical reduction generally require annealing treatment (450 ~ 550 ℃); The chemical reduction of metals usually uses solvothermal method, which requires the use of high-pressure reaction kettle. Other methods require higher temperature conditions or specific reaction equipment. Therefore, it is of great practical significance to explore a new energy-saving and convenient method for the preparation of Ti³⁺ self-doped TiO₂.

The Ti³⁺ self-doping (oxygen vacancy) effectively improves the electronic conductivity of the TiO₂ by reducing the band gap without introducing foreign atoms. However, the Ti³⁺ itself is unstable, and the surface Ti³⁺ is more easily oxidized, which leads to the difference of surface/bulk electronic structure and affects the electronic conduction of the material, so it is very important to control the concentration gradient of the Ti³⁺ and optimize its stability^[189].

3.2.3 Highly conductive material modified

The electron conduction rate of TiO₂ can be significantly increased by coating, embedding and compounding with highly conductive materials^[190]. Common highly conductive materials include elemental metals, metal fluorides, and carbon materials represented by amorphous carbon, pyrolytic carbon, carbon fibers, carbon nanotubes, graphene, MXene, and conductive polymers (polyaniline or polypyrrole)^{[88,116,191,192]}^[193]^{[151,194⇓⇓⇓ ~198]}^[199⇓~201]^[17,202]^[56,134,203]^{[55,99,204⇓⇓⇓ ~208]}^[209⇓~211]^[212].

Recently, Chen et al. Prepared a porous yolk-shell structure of N-doped carbon coated TiO₂/TiF₃ heterojunction nanobox^[193]. The TiF₃ has high electron conductivity, and the lithium storage mechanism based on the three-electron conversion reaction makes it have a high theoretical capacity of the 767 mAh·g^-1, and the lattice spacing of the perovskite structure is large, which is conducive to the diffusion of Li⁺/Na⁺. The TiO₂/TiF₃ composite structure shows good lithium and sodium storage performance, and after 100 cycles, the lithium storage capacity is 245 mAh·g^-1 (the current density is 100 mA·g^-1), and the sodium storage capacity is 112 mAh·g^-1 (the current density is 50 mA·g^-1).

Embedding TiO₂ nanocrystals uniformly in carbon matrix is an effective way to improve the extrinsic electronic conductivity and Li⁺ diffusion coefficient of TiO₂^[213,214]. Rice husk, wheat straw, sweet potato, fruit peel and hair can be used as carbon sources, and such renewable raw materials or wastes have the advantages of wide sources and low price^[215]^[216]^[217,218]^[219]^[220]. In our research group, TiO₂ nanocrystals/pyrolytic carbon composites were prepared by in-situ pyrolysis of TBOT and waste bread with spongy structure^[200]. At 100 mA·g^-1 current density, the specific capacity is 231 mAh·g^-1 after 200 cycles, corresponding to a coulombic efficiency of 99.8%, and a capacity retention of 97.5% compared to the first cycle. The results of AC impedance spectrum analysis show that the exchange current density of TiO₂/ pyrolytic carbon is 2.7×10^-3A·cm^-2, which is 9 times of that of pure TiO₂. The diffusion coefficient of the Li⁺ is two orders of magnitude higher than that of the pure TiO₂. The nanocrystalline/carbon matrix embed structure of that TiO₂ improve the interfacial charge transfer and Li⁺ diffusion capabilities.

Carbon materials can not only effectively improve electron conduction, but also significantly enhance the structural stability of composites, thus achieving ultra-long cycle life. Wang et al. Prepared ultra-long anatase TiO₂ nanofibers supported by graphene by hydrothermal method using commercial rutile TiO₂ as precursor^[204]. Graphene has good electron conductivity and can provide a smooth electron transmission channel. AC impedance measurements show that the interfacial charge transfer resistance of the TiO₂ is significantly reduced by the graphene composite. Shao et al prepared rutile TiO₂ nanofibers and rutile TiO₂/C composite nanofibers by electrospinning-high temperature annealing method using titanium isopropoxide and polyvinylpyrrolidone as raw materials^[221]. The reversible capacity of the self-supporting TiO₂/C composite nanofiber is 122 mAh·g^-1 after 100 cycles at 1 C rate. Compared with the pure TiO₂ nanofibers, the TiO₂/C composite nanofibers contain 15 wt% of carbon component, so that the composite material has higher electrical conductivity. In addition, the grain size of the TiO₂ in the TiO₂/C composite nanofiber is smaller, and the diffusion distance of the Li⁺ is shortened; The mesoporous structure of the fiber increases the contact area between the electrode and the electrolyte, and promotes the interfacial charge transfer. Li et al. Prepared TiO₂/ nitrogen-substituted carbon composite nanosheets by polymerization, carbonization and acid washing using nickel-substituted TiO₂ as template and benzidine as carbon source^[222]. The method not only reduces the strain caused by structure torsion in the carbonization process, but also prevents the collapse of the TiO₂structure caused by the insertion of the carbon layer by reasonably controlling the annealing conditions. The composite material has larger specific surface area, provides abundant active sites for ultrafast Li⁺storage, has the specific capacity of 67.8 mAh·g^-1 at the high current density of 12.8 A·g^-1, and has the capacity retention rate of 87.4% after 3000 cycles.

In order to improve the crystallinity of TiO₂, annealing treatment is usually needed in the preparation of TiO₂ materials, which is a high energy consumption process. From the point of view of energy saving, our research group proposed a flame-assisted pyrolysis method to prepare carbon-containing anatase TiO₂ in one step using ethanol solution of TBOT as precursor (Figure 8 a)^[195]. The method is simple to operate, doe not need subsequent heat treatment, and is suitable for prepare other metal oxides^[223,224]. TEM images show that the prepared TiO₂ has a microsphere morphology, and there are in-situ formed carbon elements at the grain boundaries of the TiO₂ nanocrystals (Fig. 8B, 8C), which can effectively improve the electronic conductivity of the material. Furthermore, the ball milling treatment reduces the particle size, and the obtained submicron-sized carbon-containing TiO₂ has excellent cycle stability, and the reversible capacity is 160 mAh·g^-1 after 960 charge-discharge cycles at a 335 mA·g^-1 current density.

The TiO₂/ carbon composite material combines the advantages of excellent conductivity of carbon-based materials and high safety of TiO₂ anode materials, and is most likely to be industrialized in the practical application process of TiO₂ electrodes, and is an important research direction in the field of lithium ion batteries. Table 3 summarizes the cycle performance and rate capability of some recent -carbon composites.

3.3 Phase Structure Optimization

The Li⁺ diffusion coefficient and intrinsic/extrinsic electronic conductivity of TiO₂ can be improved by designing the nanostructure and improving the electronic structure, but the lithium storage mechanism and intercalation potential range of TiO₂ with different crystal forms are different. The TiO₂ mixed crystal structure can form a heterojunction at the interface and accelerate the transfer of electrons at the interface by using the position difference of the conduction band; At the same time, the growth of specific crystal face is controlled to promote the rapid migration of Li⁺ at the phase interface.

The heterojunction interface of the anatase TiO₂/TiO₂(B) composite structure can accelerate electron transfer and Li⁺ migration, play a synergistic role in lithium storage, and contribute additional lithium storage capacity^[232⇓~234]. TiO₂(B) is a thermodynamic metastable phase, which is easy to transform into anatase TiO₂ or even rutile TiO₂ under heating conditions. The content of the TiO₂(B) in the anatase TiO₂/TiO₂(B) composite material can be adjusted by controlling the annealing temperature. When the temperature reaches 350 ℃, the TiO₂(B) begins to transform into anatase^[235]. When the temperature is in the range of 400 ~ 500 ℃, most of the TiO₂(B) are transformed into anatase phase. When the temperature is higher than 550 ℃, almost all of the TiO₂(B) transforms into anatase phase^[236]. Wu et al. Prepared microspheres composed of ultra-thin anatase TiO₂, while loading a small amount of TiO₂(B) particles inside^[234]. The contents of TiO₂(B) in the samples annealed at 450 ℃, 550 ℃ and 650 ℃ are 14.96%, 2.72% and 1.25%, respectively. The sample annealed at 550 ℃ exhibits a reversible specific capacity of 180 and 110 mAh·g^-1 after 1000 cycles at high current densities of 3. 4 and 8.5 A·g^-1, respectively, showing good cycle stability and rate capability. Lu et al. Also studied the effect of a small amount of TiO₂(B) particle modification on the improvement of lithium storage performance of anatase TiO₂^[237]. In the mesoporous anatase TiO₂/TiO₂(B) microspheres, the content of the TiO₂(B) is 5 wt%, and the TiO₂(B) is mainly distributed on the surface of the anatase TiO₂ particles. At 330 mA·g^-1 current density, the stable capacity of anatase TiO₂/TiO₂(B) composite after 80 charge-discharge cycles is 171 mAh·g^-1, which is higher than that of pure anatase TiO₂(162 mAh·g^-1) or TiO₂(B)(159 mAh·g^-1).

The band gaps of anatase TiO₂ and TiO₂(B) are both 3.2 eV, but the conduction band position of TiO₂(B) is higher than that of anatase TiO₂.The anatase TiO₂ is an electron acceptor, and electrons tend to migrate from the TiO₂(B)) to the anatase TiO₂, accelerating electron transfer at the two-phase interface (fig. 9a); Meanwhile, the TiO₂(B) has a more open pore structure in the [010] crystal direction than the anatase TiO₂, which is beneficial to the diffusion of Li⁺,The TiO₂(B) is a Li⁺ acceptor, and the Li⁺ tends to migrate from the anatase TiO₂ to the TiO₂(B) (fig. 9B). Through the construction of anatase TiO₂/TiO₂(B) heterojunction, the migration rate of Li⁺ and electrons at the interface can be improved, and the charge separation at the solid/solid interface can be accelerated. Moreover, the interface between the two phases can provide additional Li⁺ diffusion channels and active sites for lithium storage, and the kinetic process of lithium storage at the interface is faster than that in the bulk. Therefore, the existence of the anatase TiO₂/TiO₂(B) heterojunction can improve the electrochemical performance.

The construction of the anatase TiO₂/TiO₂(B) heterojunction improves the interface lithium storage activity, and in order to give full play to the lithium storage performance of the anatase TiO₂ and the TiO₂(B), nano-structures such as nanowires, nanorods, core-shells, hollow multi-shells and the like can be designed to improve the Li⁺ diffusion coefficient in a bulk phase and can be compounded with graphene to improve the electronic conductivity^[235]^[238]^[239]^[70]^{[233,236,240]}.

Compared with anatase TiO₂ and TiO₂(B), rutile TiO₂ is the most thermodynamically stable phase, and the rock salt phase formed by the first discharge is also stable and has certain lithium storage activity. The preparation and electrochemical properties of anatase TiO₂/ rutile TiO₂ bicrystal composite materials have been reported in several papers. Recently, Pan et al. Developed a method to prepare homogeneous nanostructured biphase TiO₂ porous spheres by UV-induced interfacial crystallization. By adjusting the solvent and light wavelength, anatase phase, rutile phase or biphase TiO₂ can be selectively formed on the surface of the spheres^{[241⇓⇓⇓⇓~246]}^[247]. The specific surface area of the TiO₂ formed in the mixed solvent of water and methanol under 254 nm ultraviolet light irradiation for 4 H is 358 m²·g^-1, and the lithium storage capacity reaches 128.6 mAh·g^-1 at 1 C rate for 1000 cycles. However, the lithium storage performance of rutile TiO₂ itself is not outstanding, and there is a large volume change (5% -10%) during the first irreversible phase transition, which limits the development of anatase TiO₂/ rutile TiO₂ composite structure.Currently, the main focus is to construct anatase /TiO₂(B) heterojunction and study the synergistic lithium storage characteristics of mixed crystal TiO₂^[53].

4 Performance of Li-ion battery based on TiO₂ anode material

By matching different types of cathode materials (such as polyanionic LiFePO₄, layered LiCoO₂, spinel LiMn₂O₄ and LiNi_0.5Mn_1.5O₄),TiO₂ anode materials), they have been successfully applied to lithium-ion batteries^[248⇓~250]^{[112,251,252]}^[253,254]^[255,256]. Table 4 summarizes the performance of lithium-ion full batteries based on TiO₂ anode materials. It can be seen that the full batteries matched with different cathode materials still have the shortcomings of low first coulombic efficiency and poor cycle performance.

In order to improve the practicability of TiO₂ anode materials in the whole battery, the following aspects can be optimized: (1) Material modification. The micro-nano structure, the electronic structure and the phase structure of the material are further optimized and modified to improve the structural stability, the energy density and the power density of the material, and particularly, the characteristic lithium intercalation potential platform advantage of an anatase TiO₂ with long cycle stability and the fast charging characteristic of a TiO₂(B) based on a pseudocapacitor lithium storage mechanism are fully exerted. (2) Process optimization. Secondary granulation can effectively reduce the active sites of side reactions on the surface of active materials, which is an effective means to improve the volumetric energy density, the first coulombic efficiency and the cycle stability. The development of mechanochemical and other large-scale preparation technologies and energy-saving synthesis processes such as low-temperature wet chemistry are also of great significance for practical applications. (3) Parameter matching. Negative electrode materials are prone to side reactions under too low potential conditions, which consume electrolyte and shorten the service life of the whole battery, so it is necessary to reasonably control the voltage range of the whole battery, especially the upper voltage limit. Compared with the cathode material, the first coulombic efficiency of the TiO₂ anode material is lower, which leads to a larger irreversible capacity in the first cycle of the whole battery. Therefore, it is necessary to match the appropriate cathode/anode materials, and the optimal P/N value is in the range of 1.2 to 2.0. (4) Device design. In order to avoid the potential safety hazards such as thermal runaway and spontaneous combustion caused by flammable organic electrolytes, all-solid-state batteries based on polymer electrolytes or oxide (sulfide) electrolytes are designed to further improve the safety of batteries^{[254,257⇓ ~259]}^[260].

5 Conclusion and prospect

Titanium dioxide has the advantages of non-toxicity, wide source and stable structure, and its high electrochemical working window can effectively avoid the formation of lithium dendrites, so it is a highly safe anode material for lithium-ion batteries with commercial prospects and deserves attention. At present, the modification research mainly focuses on micro/nano structure, phase structure design and electronic structure control to improve the interfacial charge transfer, intrinsic electron conduction and Li⁺ diffusion rate of TiO₂.

The two crystal forms of anatase and TiO₂(B) show high electrochemical activity and good cycle stability based on two-phase solid solution and pseudo-capacitive lithium storage mechanism, respectively, but their actual specific capacity (160~250 mAh·g^-1) is still much lower than the theoretical specific capacity (335 mAh·g^-1).The construction of anatase /TiO₂(B) mixed crystal nanostructures can give full play to the advantages of heterogeneous interface charge transfer and Li⁺ fast transmission to achieve synergistic lithium storage, which is expected to further improve the actual capacity. Ti³⁺ self-doping (oxygen vacancy) can introduce impurity energy levels, reduce the band gap, increase the carrier concentration, and enhance the intrinsic electronic conductivity.However, Ti³⁺ self-doping usually distributes on the bulk surface and is easy to be oxidized, so it is an important research direction to control the concentration gradient of Ti³⁺ and optimize its stability. Due to the low cell density of metastable TiO₂(B), the doping of ions is prone to lattice distortion, which leads to the blockage of Li⁺ diffusion channels. Therefore, it is very important to control the type and concentration of ion doping for optimizing the rate characteristics and cycle stability of TiO₂(B).

Rutile TiO₂ is the most thermodynamically stable state, and its first irreversible phase transition process has been widely recognized, but the reversible lithium storage mechanism of LiTiO₂ intermediate state in rock salt phase is still unclear, which needs to be further studied with advanced characterization techniques.

In addition to traditional lithium-ion batteries, TiO₂ materials also have good development prospects in secondary batteries, flexible devices and low-temperature application scenarios: (1) Research on alkali metal batteries such as sodium and potassium, as well as multivalent ion batteries such as magnesium and aluminum, to understand the mechanism of ion storage, establish structure-activity relationship, and develop secondary batteries with high specific energy^[261]^[262,263]^[264⇓~266]^[267⇓~269]. (2) Develop flexible secondary batteries and wearable electronic devices to expand practical applications, such as flexible display, electronic skin, wearable sensors, etc^[270⇓~272]^[273⇓~275]. (3) Give full play to the advantages of structural stability of TiO₂, and develop high-safety low-temperature batteries for extreme conditions or special application scenarios such as cold, low-latitude areas and deep space exploration.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Xie J, Lu Y C. Nat. Commun., 2020, 11(1): 1.

[2]	Zhao Y, Kang Y Q, Jin Y H, Wang L, Tian G Y, He X M. Prog. Chem., 2019, 31: 613.

[3]	(赵云, 亢玉琼, 金玉红, 王莉, 田光宇, 何向明. 化学进展, 2019, 31: 613.).

[4]	Ren W F, Zhou Y, Li J T, Huang L, Sun S G. Curr. Opin. Electrochem., 2019, 18: 46.

[5]	Fang S, Bresser D, Passerini S. Adv. Energy Mater., 2020, 10(1): 1902485.

[6]	Zhu M Y, Chen Q, Tong W J, Kan J R, Sheng W C. Prog. Chem., 2017, 29: 1366.

[7]	(朱脉勇, 陈齐, 童文杰, 阚加瑞, 盛维琛. 化学进展, 2017, 29: 1366.).

[8]	Liu X, Zhao H L, Xie J Y, Wang K, Lv P P, Gao C H. Prog. Chem., 2014, 26: 1586.

[9]	(刘欣, 赵海雷, 解晶莹, 王可, 吕鹏鹏, 高春辉. 化学进展, 2014, 26: 1586.).

[10]	Yao F, Pham D T, Lee Y H. Chem. Sus. Chem., 2015, 8: 2284.

[11]	Huang C S, Li Y L. Acta Phys. Chimica. Sin., 2016, 32: 1314.

[12]	(黄长水, (李玉良. 物理化学学报, 2016, 32: 1314.).

[13]	Li Q D, Li Y, Chen Y, Wu L L, Yang C F, Cui X L. Carbon, 2018, 136: 248.

[14]	LEE Jordan, LI Yong, TANG Jianing, CUI Xiaoli. Acta Phys. Chimica Sin., 2018, 34(9): 1080.

[15]	(李乔丹, 李永, 唐佳宁, 崔晓莉. 物理化学学报, 2018, 34(9): 1080.)

[16]	Chen Y H, Li J F, Liu H B. Acta Phys. Chimica Sin., 2018, 34(9): 1074.

[17]	(陈彦焕, 李教富, 刘辉彪. 物理化学学报, 2018, 34(9): 1074.)

[18]	Yang C F, Li Y, Chen Y, Li Q D, Wu L L, Cui X L. Small, 2019, 15(8): 1804710.

[19]	Yang C F, Qiao C, Chen Y, Zhao X Q, Wu L L, Li Y, Jia Y, Wang S Y, Cui X L. Small, 2020, 16(10): 1907365.

[20]	Zhu G N, Wang Y G, Xia Y Y. Energy Environ. Sci., 2012, 5(5): 6652.

[21]	Yuan T, Tan Z P, Ma C R, Yang J H, Ma Z F, Zheng S Y. Adv. Energy Mater., 2017, 7(12): 1601625.

[22]	Zhang Y X, Luo Y, Chen Y, Lu T L, Yan L Q, Cui X L, Xie J Y. ACS Appl. Mater. Interfaces, 2017, 9(20): 17145.

[23]	Cheng Z L, Hu Y, Wu K S, Xing Y S, Pan P, Jiang L Y, Mao J T, Ni C K, Wang Z X, Zhang M M, Zhang Y R, Gu X F, Zhang X W. Electrochimica Acta, 2020, 337: 135789.

[24]	Zhou N, Wu Y F, Li Y R, Yang J K, Zhou Q, Guo Y W, Xia M, Zhou Z. Appl. Surf. Sci., 2020, 500: 144026.

[25]	Li B Q, Zhao W, Yang Z, Zhang C, Dang F, Liu Y L, Jin F, Chen X. J. Power Sources, 2020, 466: 228339.

[26]	Fang R, Miao C, Mou H Y, Xiao W. J. Alloys Compd., 2020, 818: 152884.

[27]	Kim H, Kim M C, Choi S, Moon S H, Kim Y S, Park K W. Nanoscale, 2019, 11(37): 17415.

[28]	Liu S T, Li Y H, Zhang C, Chen X, Wang Z, Cui F M, Yang X J, Yue W B. Electrochimica Acta, 2020, 332: 135458.

[29]	Chen P, Wang Z X, Zhang B Y, Liu H, Liu W Q, Zhao J X, Ma Z H, Dong W Y, Su Z M. RSC Adv., 2020, 10(8): 4538.

[30]	Dong W J, Wang D, Li X Y, Yao Y, Zhao X, Wang Z, Wang H E, Li Y, Chen L H, Qian D, Su B L. J. Energy Chem., 2020, 48: 259.

[31]	Opra D P, Gnedenkov S V, Sinebryukhov S L. J. Power Sources, 2019, 442: 227225.

[32]	Zhang Y Y, Tang Y X, Li W L, Chen X D. ChemNanoMat, 2016, 2(8): 764.

[33]	Liu Y, Yang Y. J. Nanomater., 2016, 2016: 1.

[34]	Aravindan V, Lee Y S, Yazami R, Madhavi S. Mater. Today, 2015, 18(6): 345.

[35]	Zhang Y X, Zhang X L, Zheng H H. Battery, 2009, 39: 106.

[36]	(张玉玺, 张晓丽, 郑洪河. 电池, 2009, 39: 106.).

[37]	Vazquez-Santos M B, Tartaj P, Morales E, Amarilla J M. Chem. Rec., 2018, 18(7/8): 1178.

[38]	Yan X D, Wang Z H, He M, Hou Z H, Xia T, Liu G, Chen X B. Energy Technol., 2015, 3(8): 801.

[39]	Aravindan V, Lee Y S, Madhavi S. Adv. Energy Mater., 2015, 5(13): 1402225.

[40]	Liu Y B, Liu M Y, Lan T B, Dou J, Wei M D. J. Mater. Chem. A, 2015, 3(37): 18882.

[41]	Zhang W F, Shen D L, Liu Z W, Wu N L, Wei M D. Chem. Commun., 2018, 54(81): 11491.

[42]	Mukai K, Yamada I. J. Electrochem. Soc., 2017, 164(14): A3590.

[43]	Noailles L D, Johnson C S, Vaughey J T, Thackeray M M. J. Power Sources, 1999, 81/82: 259.

[44]	Drozhzhin O A, Grigoriev V V, Alekseeva A M, Ryazantsev S V, Tyablikov O A, Chernyshov D, Abakumov A M, Antipov E V. Eur. J. Inorg. Chem., 2020, 2020(9): 743.

[45]	Kanazawa R, Haruta M, Doi T, Inaba M. Electrochemistry, 2015, 83(10): 867.

[46]	Pérez-Flores J C, Hoelzel M, García-Alvarado F, Kuhn A. ChemPhysChem, 2016, 17(7): 1062.

[47]	Shin J Y, Samuelis D, Maier J. Adv. Funct. Mater., 2011, 21(18): 3464.

[48]	Bing Z, Yuan Y, Wang Y, Fu Z W. Electrochem. Solid-State Lett., 2006, 9(3): A101.

[49]	Lou S F, Zhao Y, Wang J J, Yin G P, Du C Y, Sun X L. Small, 2019, 15(52): 1904740.

[50]	Augustyn V, Simon P, Dunn B. Energy Environ. Sci., 2014, 7(5): 1597.

[51]	Fehse M, Ventosa E. ChemPlusChem, 2015, 80(5): 785.

[52]	Aricò A S, Bruce P, Scrosati B, Tarascon J M, van Schalkwijk W. Nat. Mater., 2005, 4(5): 366.

[53]	Zukalová M, Kalbáč M, Kavan L, Exnar I, Graetzel M. Chem. Mater., 2005, 17(5): 1248.

[54]	Zhang Q, Kaghazchi P. J. Phys. Chem. C, 2016, 120(39): 22163.

[55]	Lützenkirchen-Hecht D, Wagemaker M, Keil P, van Well A A, Frahm R. Surf. Sci., 2003, 538(1/2): 10.

[56]	Hu Y S, Kienle L, Guo Y G, Maier J. Adv. Mater., 2006, 18(11): 1421.

[57]	Kim S J, Noh S Y, Kargar A, Wang D L, Graham G W, Pan X Q. Chem. Commun., 2014, 50(69): 9932.

[58]	Pan X Q, Kim S J, Zhang K, Jokisaari J R, Kargar A, Wang D, Graham G W. Microsc. Microanal., 2015, 21(S3): 1367.

[59]	Baudrin E, Cassaignon S, Koelsch M, Jolivet J P, Dupont L, Tarascon J M. Electrochem. Commun., 2007, 9(2): 337.

[60]	Christensen C K, Mamakhel M A H, Balakrishna A R, Iversen B B, Chiang Y M, Ravnsbæk D B. Nanoscale, 2019, 11(25): 12347.

[61]	Kubiak P, Pfanzelt M, Geserick J, Hörmann U, Hüsing N, Kaiser U, Wohlfahrt-Mehrens M. J. Power Sources, 2009, 194(2): 1099.

[62]	Mo R W, Lei Z Y, Sun K N, Rooney D. Adv. Mater., 2014, 26(13): 2084.

[63]	Tang Y K, Liu L, Wang X C, Jia D Z, Xia W, Zhao Z B, Qiu J S. J. Power Sources, 2016, 319: 227.

[64]	Di Lupo F, Tuel A, Mendez V, Francia C, Meligrana G, Bodoardo S, Gerbaldi C. Acta Mater., 2014, 69: 60.

[65]	Li X J, Li M Y, Liang J C, Wang X F, Yu K F. J. Alloys Compd., 2016, 681: 471.

[66]	Lan T B, Tu J X, Zou Q M, Zeng X R, Zou J Z, Huang H T, Wei M D. Electrochimica Acta, 2019, 319: 101.

[67]	Ren Y, Liu Z, Pourpoint F, Armstrong A R, Grey C P, Bruce P G. Angew. Chem. Int. Ed., 2012, 51(9): 2164.

[68]	Kashale A A, Dwivedi P K, Sathe B R, Shelke M V, Chang J Y, Ghule A V. ACS Omega, 2018, 3: 13676.

[69]	Li C, Zhao M, Sun C N, Jin B, Yang C C, Jiang Q. J. Power Sources, 2018, 397: 162.

[70]	El Ouardi K, Dahbi M, Hakim C, Güler M O, Akbulut H, El Bouari A, Saadoune I. J. Appl. Electrochem., 2020, 50(5): 583.

[71]	Zhang P G, Zhang C Y, Xie A J, Li C, Song J M, Shen Y H. J. Mater. Sci., 2016, 51(7): 3448.

[72]	Liu H, Li W, Shen D K, Zhao D Y, Wang G X. J. Am. Chem. Soc., 2015, 137(40): 13161.

[73]	Shi S, Chen Y, Lee J, Jiang Z Y, Cui X L. J. Solid State Electrochem., 2018, 22(3): 705.

[74]	Dylla A G, Lee J A, Stevenson K J. Langmuir, 2012, 28(5): 2897.

[75]	Wang S, Yu X H, Liu J X, Dong P, Zhang Y J, Zhu C Y, Zhan Z L, Zhang Y N. J. Alloys Compd., 2020, 814: 152342.

[76]	Hou J, Zhang H M, Lin J J, Qiu X Y, Zhao W S, Sun X G, Xiang Y, Zhang H, Xing G C, Zheng D H, Li G D, Tang Z Y. J. Mater. Chem. A, 2019, 7(41): 23733.

[77]	Ren H, Yu R B, Qi J, Zhang L J, Jin Q, Wang D. Adv. Mater., 2019, 31(10): 1805754.

[78]	Ren H, Yu R B, Wang J Y, Jin Q, Yang M, Mao D, Kisailus D, Zhao H J, Wang D. Nano Lett., 2014, 14(11): 6679.

[79]	Zhang G Q, Wu H B, Song T, Paik U, Lou X W D. Angew. Chem., 2014, 126(46): 12798.

[80]	Zhang Y F, Zhang N, Chen J, Zhang T Z, Ge W Q, Zhang W M, Xie G, Zhang L P, He Y H. J. Alloys Compd., 2020, 815: 152511.

[81]	Salman M S, Park A R, Cha M J, Choi Y, Jang S K, Tan L H, Yoo P J, Choe W S. ACS Appl. Nano Mater., 2018, 1(2): 698.

[82]	Liu R P, Shen C, Zhang C, Iocozzia J, Wang Q, Zhao S Q, Yuan K J, Lin Z Q. J. Mater. Sci., 2018, 53(11): 8499.

[83]	Long L Y, Zhang H, Ye M, Fang Z. RSC Adv., 2015, 5(16): 12224.

[84]	Wang X B, Meng Q X, Wang Y Y, Liang H J, Bai Z Y, Wang K, Lou X D, Cai B B, Yang L. Appl. Energy, 2016, 175: 488.

[85]	Han C P, Yang D, Yang Y K, Jiang B B, He Y J, Wang M Y, Song A Y, He Y B, Li B H, Lin Z Q. J. Mater. Chem. A, 2015, 3(25): 13340.

[86]	Wang X B, Wang Y Y, Yang L, Wang K, Lou X D, Cai B B. J. Power Sources, 2014, 262: 72.

[87]	Yu K F, Ling M Q, Liang J C, Liang C. Chem. Phys., 2019, 517: 222.

[88]	Li L, Zhang J, Zou Y L, Jiang W J, Lei W X, Ma Z S. J. Electroanal. Chem., 2019, 833: 573.

[89]	Zhang W F, Zhang Y, Yu L, Wu N L, Huang H T, Wei M D. J. Mater. Chem. A, 2019, 7(8): 3842.

[90]	Wang Y, Zhang J. Ionics, 2020, 26(3): 1159.

[91]	Li Y H, Tang X, Zhou X J, Li L, Jiang S. Appl. Surf. Sci., 2020, 505: 144649.

[92]	Bao S J, Bao Q L, Li C M, Dong Z L. Electrochem. Commun., 2007, 9(5): 1233.

[93]	Giannuzzi R, Manca M de Marco L, Belviso M R, Cannavale A, Sibillano T, Giannini C, Cozzoli P D, Gigli G. ACS Appl. Mater. Interfaces, 2014, 6(3): 1933.

[94]	Wen W, Wu J M, Jiang Y Z, Yu S L, Bai J Q, Cao M H, Cui J. Sci. Rep., 2015, 5(1): 1.

[95]	He B L, Dong B, Li H L. Electrochem. Commun., 2007, 9(3): 425.

[96]	Brutti S, Gentili V, Menard H, Scrosati B, Bruce P G. Adv. Energy Mater., 2012, 2(3): 322.

[97]	Qu J, Cloud J E, Yang Y, Ding J N, Yuan N Y. ACS Appl. Mater. Interfaces, 2014, 6(24): 22199.

[98]	Tang Y X, Zhang Y Y, Deng J Y, Wei J Q, Tam H L, Chandran B K, Dong Z L, Chen Z, Chen X D. Adv. Mater., 2014, 26(35): 6111.

[99]	Huo J H, Xue Y J, Liu Y, Guo S W. J. Electroanal. Chem., 2020, 863: 114088.

[100]

Smith K A, Savva A I, Mao K S, Wang Y Q, Tenne D A, Chen D, Liu Y Z, Barnes P, Deng C J, Butt D P, Wharry J P, Xiong H. J. Mater. Sci., 2019, 54(20): 13221.

[101]

Dong S M, Wang H B, Gu L, Zhou X H, Liu Z H, Han P X, Wang Y, Chen X, Cui G L, Chen L Q. Thin Solid Films, 2011, 519(18): 5978.

[102]

Wang Q W, Du X F, Chen X Z, Xu Y L. Acta Phys. Chimica. Sin., 2015, 31: 1437.

[103]

(汪倩雯, 杜显锋, 陈夕子, 徐友龙. 物理化学学报, 2015, 31: 1437.).

[104]

Chen J S, Lou X W. Electrochem. Commun., 2009, 11(12): 2332.

[105]

Liu S H, Jia H P, Han L, Wang J L, Gao P F, Xu D D, Yang J, Che S N. Adv. Mater., 2012, 24(24): 3201.

[106]

Zhang W, Zhou W D, Wright J H, Kim Y N, Liu D W, Xiao X C. ACS Appl. Mater. Interfaces, 2014, 6(10): 7292.

[107]

Liang Y R, Xiong X, Xu Z J, Xia Q B, Wan L Y, Liu R T, Chen G X, Chou S L. Small, 2020, 16(26): 2000030.

[108]

Kim G, Jo C, Kim W, Chun J, Yoon S, Lee J, Choi W. Energy Environ. Sci., 2013, 6(10): 2932.

[109]

Huang X K, Cui S M, Wieboldt R C, Hallac P B, Fell C R, Metz B, Jiang J W, Chen J H. Adv. Electron. Mater., 2015, 1(12): 1500256.

[110]

Zhu L P, Huang C, Liu J W, Bing N C, Jin H Y, Wang L J. Mater. Lett., 2013, 106: 348.

[111]

Yu X Y, Wu H B, Yu L, Ma F X, Lou X W D. Angew. Chem. Int. Ed., 2015, 54(13): 4001.

[112]

Choi T, Kim S D, Yeo S, Cheon T, Kim S H, Ahn J H, Kim H. Electrochimica Acta, 2020, 334: 135596.

[113]

Auer A, Kunze-Liebhäuser J. Small Methods, 2019, 3(8): 1800385.

[114]

Liu Y, Yan X D, Xu B Q, Lan J L, Yu Y H, Yang X P, Lin Y H, Nan C W. ACS Appl. Mater. Interfaces, 2018, 10(22): 19047.

[115]

Hu H, Yu L, Gao X H, Lin Z, Lou X W D. Energy Environ. Sci., 2015, 8(5): 1480.

[116]

Wu N T, Qiao X G, Shen J K, Liu G L, Sun T, Wu H, Hou H S, Liu X M, Zhang Y, Ji X B. Electrochimica Acta, 2019, 299: 540.

[117]

Choi S I, Jung E J, Park M, Shin H S, Huh S, Won Y S. Appl. Surf. Sci., 2020, 508: 145237.

[118]

Ren H, Yu R B. Chin. Sci. Bull., 2019, 64: 3546.

[119]

(任浩, 于然波. 科学通报, 2019, 64: 3546.).

[120]

McNulty D, Carroll E, O'Dwyer C. Adv. Energy Mater., 2017, 7(12): 1602291.

[121]

Dong C, Li A, Zhang L G, Dong W J, Gao H Y, Jia X L, Chen X, Mi H T, Wang G, Chen X B. J. Power Sources, 2020, 448: 227458.

[122]

Xie K Y, Guo M, Lu W, Huang H T. Nanotechnology, 2014, 25(45): 455401.

[123]

Liu X, Sun Q, Ng A M C, Djurišić A B, Xie M H, Liao C Z, Shih K, Vranješ M, Nedeljković J M, Deng Z F. Nanotechnology, 2015, 26(42): 425403.

[124]

Li X, Wu G, Liu X, Li W, Li M. Nano Energy, 2017, 31: 1.

[125]

Yin J P, Wen Z S, Yu J Y, Shi X R, Wang G Q, Yang Y E, Man J Z, Sun J C, Cui L Y. J. Alloys Compd., 2019, 808: 151728.

[126]

Zhao Y, Jin Y H, Wang L, Tian G Y, He X M. Prog. Chem., 2018, 30: 1761.

[127]

(赵云, 金玉红, 王莉, 田光宇, 何向明. 化学进展, 2018, 30: 1761.).

[128]

Gao R M, Jiao Z, Wang Y, Xu L Q, Xia S S, Zhang H J. Chem. Eng. J., 2016, 304: 156.

[129]

Chen J S, Liang Y N, Li Y M, Yan Q Y, Hu X. ACS Appl. Mater. Interfaces, 2013, 5(20): 9998.

[130]

Qiao H, Luo Q H, Wei Q F, Cai Y B, Huang F L. Ionics, 2012, 18(7): 667.

[131]

Bai Y, Liu Z M, Zhang N Q, Sun K N. RSC Adv., 2015, 5(27): 21285.

[132]

Wang D D, Shan Z Q, Na R, Huang W L, Tian J H. J. Power Sources, 2017, 337: 11.

[133]

Jiang C H, Honma I, Kudo T, Zhou H S. Electrochem. Solid-State Lett., 2007, 10(5): A127.

[134]

Dylla A G, Henkelman G, Stevenson K J. Acc. Chem. Res., 2013, 46(5): 1104.

[135]

Nara H, Morita K, Mukoyama D, Yokoshima T, Momma T, Osaka T. Electrochimica Acta, 2017, 241: 323.

[136]

Qin G H, Ma Q Q, Wang C Y. Solid State Ion., 2014, 257: 60.

[137]

Wang J Y, Liao L, Li Y Z, Zhao J, Shi F F, Yan K, Pei A, Chen G X, Li G D, Lu Z Y, Cui Y. Nano Lett., 2018, 18(11): 7060.

[138]

Xu X Q, Wan Y X, Zuo C, Yan H P, Du M T, Xue G, Zhou D S. Mater. Today Energy, 2018, 7: 80.

[139]

Takami N, Harada Y, Iwasaki T, Hoshina K, Yoshida Y. J. Power Sources, 2015, 273: 923.

[140]

Jo C, Groombridge A S, de la Verpilliere J, Lee J T, Son Y, Liang H L, Boies A M, de Volder M. ACS Nano, 2020, 14(1): 698.

[141]

Lee P K, Tan T, Wang S, Kang W P, Lee C S, Yu D Y W. ACS Appl. Mater. Interfaces, 2018, 10(40): 34132.

[142]

Xun S D, Xiang B, Minor A, Battaglia V, Liu G. J. Electrochem. Soc., 2013, 160(9): A1380.

[143]

Kashale A A, Rasal A S, Kamble G P, Ingole V H, Dwivedi P K, Rajoba S J, Jadhav L D, Ling Y C, Chang J Y, Ghule A V. Compos. B: Eng., 2019, 167: 44.

[144]

Ata-Ur-rehman , Ali G, Abbas S M, Iftikhar M, Zahid M, Yaseen S, Saleem S, Haider S, Arshad M, Badshah A. Chem. Eng. J., 2019, 375: 122021.

[145]

Ali Z, Cha S N, Sohn J I, Shakir I, Yan C Z, Kim J M, Kang D J. J. Mater. Chem., 2012, 22(34): 17625.

[146]

Zhang C, Qi Y C, Liu S T, Men Y, Cui F M. Mater. Chem. Phys., 2019, 237: 121822.

[147]

Xu H, Zeng M, Li J, Li F Y. Nano, 2016, 11(1): 1650006.

[148]

Bi Z H, Paranthaman M P, Guo B K, Unocic R R, Meyer H M, Bridges C A, Sun X G, Dai S. J. Mater. Chem. A, 2014, 2(6): 1818.

[149]

Grosjean R, Fehse M, Pigeot-Remy S, Stievano L, Monconduit L, Cassaignon S. J. Power Sources, 2015, 278: 1.

[150]

Fang J, Liu W W, Yu F, Qin F R, Wang M R, Zhang K, Lai Y Q. RSC Adv., 2016, 6(74): 70133.

[151]

Abhilash K P, Christopher Selvin P, Nalini B, Jose R, Vijayaraghavan R, Chowdari B V R, Adams S, Reddy M V. J. Alloys Compd., 2017, 710: 205.

[152]

Wang Y L, Xu M, Peng Z, Zheng G F. J. Mater. Chem. A, 2013, 1(42): 13222.

[153]

Lübke M, Johnson I, Makwana N M, Brett D, Shearing P, Liu Z L, Darr J A. J. Power Sources, 2015, 294: 94.

[154]

Jiao S L, Lian G, Jing L Y, Xu Z H, Wang Q L, Cui D L, Wong C P. ACS Omega, 2018, 3(1): 1329.

[155]

Reddy M V, Sharma N, Adams S, Rao R P, Peterson V K, Chowdari B V R. RSC Adv., 2015, 5(37): 29535.

[156]

Opra D P, Gnedenkov S V, Sinebryukhov S A, Ustinov A Y, Podgorbunsky A B, Sokolov A A. Russ. J. Inorg. Chem., 2019, 64(5): 680.

[157]

Yang H, Lan C K, Duh J G. J. Power Sources, 2015, 288: 401.

[158]

Zhao F, Wang B F, Tang Y F, Ge H H, Huang Z G, Liu H K. J. Mater. Chem. A, 2015, 3(45): 22969.

[159]

Usui H, Domi Y, Yoshioka S, Kojima K, Sakaguchi H. ACS Sustainable Chem. Eng., 2016, 4(12): 6695.

[160]

Lan T B, Zhang W F, Wu N L, Wei M D. Chem. Eur. J., 2017, 23(21): 5059.

[161]

Xia Y, Rong C, Yang X, Lu F, Kuang X. Front. Mater., 2019, 6: 1.

[162]

Cao M L, Tao L M, Lv X, Bu Y, Li M, Yin H, Zhu M Q, Zhong Z C, Shen Y, Wang M K. J. Power Sources, 2018, 396: 327.

[163]

Tian H J, Xin F X, Tan X J, Han W Q. J. Mater. Chem. A, 2014, 2(27): 10599.

[164]

Jung H G, Yoon C S, Prakash J, Sun Y K. J. Phys. Chem. C, 2009, 113(50): 21258.

[165]

Zhang Y Q, Du F, Yan X, Jin Y M, Zhu K, Wang X, Li H M, Chen G, Wang C Z, Wei Y J. ACS Appl. Mater. Interfaces, 2014, 6(6): 4458.

[166]

Zhang Y Q, Fu Q, Xu Q L, Yan X, Zhang R Y, Guo Z D, Du F, Wei Y J, Zhang D, Chen G. Nanoscale, 2015, 7(28): 12215.

[167]

Choi W H, Lee C H, Kim H E, Lee S U, Bang J H. Nano Energy, 2020, 74: 104829.

[168]

Gao Y P, Wang C, Hu P F, He F Y, Wu M H, Zhang H J. J. Alloys Compd., 2019, 787: 944.

[169]

Gao D L, Wang Y L, Kong J, Huo F, Wang S F, He H Y, Zhang S J. Phys. Chem. Chem. Phys., 2019, 21(32): 17985.

[170]

Lai C, Yuan X C, Cao X L, Qiao Q Q, Wang Y L, Ye S H. Electrochem. Solid-State Lett., 2012, 15(5): A65.

[171]

Sun S M, Wu J, Watanabe M, Akbay T, Ishihara T. J. Phys. Chem. Lett., 2019, 10(11): 2998.

[172]

Chen S C, Wang H, Kang Z X, Jin S, Zhang X D, Zheng X S, Qi Z M, Zhu J F, Pan B C, Xie Y. Nat. Commun., 2019, 10(1): 1.

[173]

Wu Z J, Guo K, Cao S, Yao W Q, Piao L Y. Nano Res., 2020, 13(2): 551.

[174]

Feng H F, Xu Z F, Ren L, Liu C, Zhuang J C, Hu Z P, Xu X, Chen J, Wang J O, Hao W C, Du Y, Dou S X. ACS Catal., 2018, 8(5): 4288.

[175]

Li H Y, Ren F Z, Liu J F, Wang Q L, Li Q Y, Yang J J, Wang Y X. Appl. Catal. B: Environ., 2015, 172/173: 37.

[176]

Li Y H, Wu X P, Liu C, Wang M, Song B T, Yu G Y, Yang G, Hou W H, Gong X Q, Peng L M. Acta Phys. Chimica Sin., 2020, 36(4): 92.

[177]

(李玉红, 吴新平, 刘聪, 王萌, 宋本腾, 郁桂云, 杨刚, 侯文华, 龚学庆, 彭路明. 物理化学学报, 2020, 36(4): 92.)

[178]

Wang X T, Li Y M, Liu X, Gao S M, Huang B B, Dai Y. Chin. J. Catal., 2015, 36(3): 389.

[179]

Pinna N, Hochepied J F, Niederberger M, Gregg M. Phys. Chem. Chem. Phys., 2009, 11: 3607.

[180]

Zheng J, Ji G B, Zhang P, Cao X Z, Wang B Y, Yu L H, Xu Z C. Chem. Eur. J., 2015, 21(50): 18309.

[181]

Chen J, Song W X, Hou H S, Zhang Y, Jing M J, Jia X N, Ji X B. Adv. Funct. Mater., 2015, 25(43): 6793.

[182]

Kang S H, Jo Y N, Prasanna K, Santhoshkumar P, Joe Y C, Vediappan K, Gnanamuthu R, Lee C W. J. Ind. Eng. Chem., 2019, 71: 177.

[183]

Lee J, Cui X L. Mater. Lett., 2016, 175: 114.

[184]

Eom J Y, Lim S J, Lee S M, Ryu W H, Kwon H S. J. Mater. Chem. A, 2015, 3(21): 11183.

[185]

Zhang Z H, Zhou Z F, Nie S, Wang H H, Peng H R, Li G C, Chen K Z. J. Power Sources, 2014, 267: 388.

[186]

Byeon A, Boota M, Beidaghi M, Aken K V, Lee J W, Gogotsi Y. Electrochem. Commun., 2015, 60: 199.

[187]

Zheng J, Liu L, Ji G B, Yang Q F, Zheng L R, Zhang J. ACS Appl. Mater. Interfaces, 2016, 8(31): 20074.

[188]

Yang Y C, Shi W, Liao S J, Zhang R H, Leng S L. J. Alloys Compd., 2018, 746: 619.

[189]

Zhang M M, Wang C R, Li H, Wang J L, Li M, Chen X S. Electrochimica Acta, 2019, 326: 134972.

[190]

Sun L J, Liu W, Wu R T, Cui Y P, Zhang Y, Du Y X, Liu S, Liu S, Wang H L. Nanoscale, 2020, 12(2): 746.

[191]

Qiu J X, Li S, Gray E, Liu H W, Gu Q F, Sun C H, Lai C, Zhao H J, Zhang S Q. J. Phys. Chem. C, 2014, 118(17): 8824.

[192]

Zheng J, Liu Y S, Ji G B, Zhang P, Cao X Z, Wang B Y, Zhang C H, Zhou X G, Zhu Y, Shi D N. ACS Appl. Mater. Interfaces, 2015, 7(42): 23431.

[193]

Lee T Y, Lee C Y, Chiu H T. ACS Omega, 2019, 4(14): 16217.

[194]

Hao Z K, Chen Q, Dai W R, Ren Y J, Zhou Y, Yang J L, Xie S J, Shen Y B, Wu J H, Chen W, Xu G Q. Adv. Energy Mater., 2020, 10(10): 1903107.

[195]

Lee J, Li Z, Zhu L Z, Xie S H, Cui X L. Appl. Catal. B: Environ., 2018, 224: 715.

[196]

Liu Y, Liu C J, Li J L. J. Mater. Chem. A, 2014, 2(38): 15746.

[197]

Zhang W F, Xu T, Liu Z W, Wu N L, Wei M D. Chem. Commun., 2018, 54(12): 1413.

[198]

Chen X, Zhao D X, Liu K W, Wang C R, Liu L, Li B H, Zhang Z Z, Shen D Z. ACS Appl. Mater. Interfaces, 2015, 7(29): 16070.

[199]

Fan C Y, Chen C, Wang J, Fu X X, Ren Z M, Qian G D, Wang Z Y. Sci. Rep., 2015, 5(1): 1.

[200]

Zhang J W, Zhang L L, Zhang J W, Zhang Z J, Wu Z S. J. Alloys Compd., 2015, 642: 28.

[201]

Wang Y, Li N, Hou C X, He B, Li J J, Dang F, Wang J, Fan Y Q. Ceram. Int., 2020, 46(7): 9119.

[202]

Liu Q, Hou J G, Xu C X, Chen Z Z, Qin R, Liu H. Chem. Eng. J., 2020, 381: 122649.

[203]

Li Y X, Zhu S L, Inoue A, Liang Y Q, Chang C T, Luo S Y, Cui Z D. Ionics, 2020, 26(9): 4363.

[204]

Xu H, Wang W J, Yu G X, Qin L G, Jiang Y Q, Ren L H, Chen J. ACS Appl. Energy Mater., 2020, 3(5): 4738.

[205]

Yuan Y F, Chen F, Cai G C, Yin S M, Zhu M, Wang L N, Yang J L, Guo S Y. Electrochimica Acta, 2019, 296: 669.

[206]

Chen Y, Ma X Q, Cui X L, Jiang Z Y. J. Power Sources, 2016, 302: 233.

[207]

Shen X Y, Chen M, Hong X H, Wang W D, Qiao Z K, Chen J, Fan S J, Yu J X, Tang C J. Chem. Phys., 2020, 530: 110639.

[208]

Tian Y X, Xu Y H, Wang Y, Zhang J H, Yang H Z, Sun T. J. Phys. Chem. Solids, 2020, 143: 109524.

[209]

Chen Y, Song C Y, Li Z F, Cui X L, Jiang Z Y. Mater. Lett., 2017, 187: 114.

[210]

Guan Z, Wang X X, Li T T, Zhu Q Z, Jia M Q, Xu B. J. Mater. Sci. Technol., 2019, 35(9): 1977.

[211]

Chen Y, Li Z F, Shi S, Song C Y, Jiang Z Y, Cui X L. J. Mater. Sci.: Mater. Electron., 2017, 28(13): 9206.

[212]

Luo H, Chen Y X, Huang J, Chen Z L, Xia X H, Li J, Liu H B. Nanoscale, 2020, 12(13): 7366.

[213]

Xie S M, Yao T H, Wang J K, Alsulami H, Kutbi M A, Wang H K. ChemistrySelect, 2020, 5(11): 3225.

[214]

Liu X H, Zhao L J, Wang S T, Chao M, Li Y T, Leng J, Zhang J Y, Tang Z L. Sci. Bull., 2019, 64(16): 1148.

[215]

Gu G F, Cheng J L, Li X D, Ni W, Guan Q, Qu G X, Wang B. J. Mater. Chem. A, 2015, 3(12): 6642.

[216]

Zhao X N, Liu H, Ding M, Feng Y Q. Ceram. Int., 2019, 45(9): 12476.

[217]

Cheng L L, Qiao D D, Zhao P D, He Y C, Sun W F, Yu H C, Jiao Z. Electrochimica Acta, 2019, 300: 417.

[218]

Su B L, Van Tendeloo G, Peng D L, Xu B X, Hasan T, Mai L Q, Gao H X, Wang H E, Li Y, Yan M, Yi M, Jin J, Hu Z Y, Yu W B. Natl. Sci. Rev., 2020, 7: 1046.

[219]

Gong L, Yang R, Liu R, Chen L P, Yan Y L, Feng Z F, Prog. Chem., 2019, 31: 1020.

[220]

(龚乐, 杨蓉, 刘瑞, 陈利萍, 燕映霖, 冯祖飞. 化学进展, 2019, 31: 1020.).

[221]

Jia L, Li Y L, Su L J, Liu D Q, Fu Y J, Li J S, Yan X B, He D Y. ChemistrySelect, 2020, 5(10): 3124.

[222]

Li L, Jiang G X, An C H, Xie Z J, Wang Y J, Jiao L F, Yuan H T. Nanoscale, 2020, 12(18): 10369.

[223]

Jiang M, Zhang F Z, Zhu G J, Ma Y Y, Luo W, Zhou T F, Yang J P. ACS Appl. Mater. Interfaces, 2020, 12(22): 24796.

[224]

Lai C, Li G R, Dou Y Y, Gao X P. Electrochimica Acta, 2010, 55(15): 4567.

[225]

Chang P Y, Huang C H, Doong R A. Carbon, 2012, 50(11): 4259.

[226]

Wang L, Nie Z Y, Cao C B, Khalid S, Wu Y, Xu X Y. J. Power Sources, 2016, 302: 259.

[227]

Wang L P, Schnepp Z, Titirici M M. J. Mater. Chem. A, 2013, 1(17): 5269.

[228]

Chen F, Yang J, Bai T, Long B, Zhou X Y. J. Electroanal. Chem., 2016, 768: 18.

[229]

Lotfabad E M, Ding J, Cui K, Kohandehghan A, Kalisvaart W P, Hazelton M, Mitlin D. ACS Nano, 2014, 8(7): 7115.

[230]

Zheng P, Liu T, Zhang J Z, Zhang L F, Liu Y, Huang J F, Guo S W. RSC Adv., 2015, 5(51): 40737.

[231]

Sun X L, Wang X H, Feng N, Qiao L, Li X W, He D Y. J. Anal. Appl. Pyrolysis, 2013, 100: 181.

[232]

Saravanan K, Kalaiselvi N. Carbon, 2015, 81: 43.

[233]

Zhao B T, Cai R, Jiang S M, Sha Y J, Shao Z P. Electrochimica Acta, 2012, 85: 636.

[234]

Li Y T, Chen M S, Cheng J F, Fu W W, Hu Y J, Liu B H, Zhang M, Shen Z R. Langmuir, 2020, 36(9): 2255.

[235]

Ma X Q, Chen Y, Li H, Cui X L, Lin Y H. Mater. Res. Bull., 2015, 66: 51.

[236]

Li J, Chen X, Sun M X, Cui X L. Mater. Lett., 2013, 110: 245.

[237]

Gholami M, Zarei-Jelyani M, Babaiee M, Baktashian S, Eqra R. Ionics, 2020, 26(9): 4391.

[238]

Han M S, Mu Y B, Yuan F, Bai X D, Yu J. J. Power Sources, 2020, 465: 228206.

[239]

Fu W W, Li Y T, Chen M S, Hu Y J, Liu B H, Zhang K, Zhan C Y, Zhang M, Shen Z R. J. Power Sources, 2020, 468: 228363.

[240]

Zheng Y Q, Yuan Y F, Tong Z W, Yin H, Yin S M, Guo S Y. Nanotechnology, 2020, 31(21): 215407.

[241]

Subaşı Y, Somer M, Yağcı M B, Slabon A, Afyon S. J. Solid State Electrochem., 2020, 24(5): 1085.

[242]

Madhusanka S A D R. Int. J. Electrochem. Sci., 2020: 2792.

[243]

Fu Z G, Li R X, Kan Z. J. Solid State Electrochem., 2019, 23(6): 1779.

[244]

Liu G L, Wu H H, Meng Q Q, Zhang T, Sun D, Jin X Y, Guo D L, Wu N T, Liu X M, Kim J K. Nanoscale Horiz., 2020, 5(1): 150.

[245]

Li X L, Liu Y, Zhang X L, Yao C, Wang R H, Xu C H, Lei J. Electrochimica Acta, 2019, 298: 14.

[246]

Wu Q L, Xu J G, Yang X F, Lu F Q, He S M, Yang J L, Fan H J, Wu M M. Adv. Energy Mater., 2015, 5(7): 1401756.

[247]

Li K K, Li B H, Wu J X, Kang F Y, Kim J K, Zhang T Y. ACS Appl. Mater. Interfaces, 2017, 9(41): 35917.

[248]

Ni L, Wang R W, Fu Y F, Wang H B, Liu W B, Zhu L K, Qiu S L, Zhang Z T. J. Alloys Compd., 2019, 790: 683.

[249]

Zhuang W, Lu L H, Li W, An R, Feng X, Wu X B, Zhu Y D, Lu X H. Chin. J. Chem. Eng., 2015, 23(3): 583.

[250]

Gao M, Bao Y B, Qian Y X, Deng Y F, Li Y W, Chen G H. Inorg. Chem., 2018, 57(19): 12245.

[251]

Wei H, Rodriguez E F, Hollenkamp A F, Bhatt A I, Chen D H, Caruso R A. Adv. Funct. Mater., 2017, 27(46): 1703270.

[252]

Song W X, Chen J, Ji X B, Zhang X M, Xie F, Riley D J. J. Mater. Chem. A, 2016, 4(22): 8762.

[253]

Wang H, Yang H X, Lu L. RSC Adv., 2014, 4(82): 43346.

[254]

Wei J, Liu J X, Wu Z Y, Zhan Z L, Shi J, Xu K. J. Nanosci. Nanotechnol., 2015, 15(7): 5013.

[255]

Tang Y K, Liu L, Zhao H Y, Jia D Z, Xie X L, Zhang Y, Li X H. CrystEngComm, 2016, 18(24): 4489.

[256]

Liu Y B, Ding T L, Shen D L, Dou J, Wei M D. J. Electroanal. Chem., 2017, 804: 87.

[257]

Jia C C, Zhang X, Yang P. Int. J. Hydrog. Energy, 2018, 43(4): 2237.

[258]

Tao W, Wang M K, Zhu B, Huo W R, Yang R Q, Xiong H J, Tang H, Wei Z H, Wang Y. Electrochimica Acta, 2020, 334: 135569.

[259]

Ma D W, Li K M, Pan J H. ACS Appl. Energy Mater., 2020, 3(5): 4186.

[260]

Lin X J, Wang Y M, Chai W, Liu T, Mou J R, Liu J, Huang J L, Liu M L. Electrochimica Acta, 2019, 308: 253.

[261]

Zhang X, Aravindan V, Kumar P S, Liu H, Sundaramurthy J, Ramakrishna S, Madhavi S. Nanoscale, 2013, 5(13): 5973.

[262]

Guo Z Y, Dong X L, Zhou D D, Du Y J, Wang Y G, Xia Y Y. RSC Adv., 2013, 3(10): 3352.

[263]

Gangaja B, Jayasree S S, Nair S, Santhanagopalan D. ChemistrySelect, 2018, 3(43): 12258.

[264]

Kim D S, Chung D J, Bae J, Jeong G, Kim H. Electrochimica Acta, 2017, 258: 336.

[265]

Zhu K L, Li Q, Xue Z M, Yu Q, Liu X C, Shan Z Q, Liu K. ACS Appl. Nano Mater., 2020, 3(2): 1019.

[266]

Aravindan V, Shubha N, Cheah Y L, Prasanth R, Chuiling W, Prabhakar R R, Madhavi S. J. Mater. Chem. A, 2013, 1(2): 308.

[267]

Ulissi U, Zimmermann J, Brutti S, Hassoun J. Electrochimica Acta, 2016, 201: 158.

[268]

Brutti S, Gentili V, Reale P, Carbone L, Panero S. J. Power Sources, 2011, 196(22): 9792.

[269]

Armstrong G, Armstrong A R, Bruce P G, Reale P, Scrosati B. Adv. Mater., 2006, 18(19): 2597.

[270]

Chen Y, Shi Y, Liang Y L, Dong H, Hao F, Wang A, Zhu Y X, Cui X L, Yao Y. ACS Appl. Energy Mater., 2019, 2(3): 1608.

[271]

Shi Y, Chen Y, Liang Y L, Andrews J, Dong H, Yuan M Y, Ding W Y, Banerjee S, Ardebili H, Robertson M L, Cui X L, Yao Y. J. Mater. Chem. A, 2019, 7(34): 19691.

[272]

Hao F, Han F D, Liang Y L, Wang C S, Yao Y. MRS Bull., 2018, 43(10): 775.

[273]

Wang N N, Chu C X, Xu X, Du Y, Yang J, Bai Z C, Dou S X. Adv. Energy Mater., 2018, 8(27): 1801888.

[274]

Li Y P, Yang C H, Zheng F H, Pan Q C, Liu Y Z, Wang G, Liu T Z, Hu J H, Liu M L. Nano Energy, 2019, 59: 582.

[275]

Cai J S, Cai R, Sun Z T, Wang X G, Wei N, Xu F, Shao Y L, Gao P, Dou S X, Sun J Y. Nano Micro Lett., 2020, 12(1): 1.

[276]

Zhang M H, MacRae A C, Liu H D, Meng Y S. J. Electrochem. Soc., 2016, 163(10): A2368.

[277]

Wang Y R, Xue X L, Liu P Y, Wang C X, Yi X, Hu Y, Ma L B, Zhu G Y, Chen R P, Chen T, Ma J, Liu J, Jin Z. ACS Nano, 2018, 12(12): 12492.

[278]

Le T S, Hoa T H, Truong D Q. J. Electroanal. Chem., 2019, 848: 113293.

[279]

Kazazi M, Abdollahi P, Mirzaei-Moghadam M. Solid State Ion., 2017, 300: 32.

[280]

Das S K, Palaniselvam T, Adelhelm P. Solid State Ion., 2019, 340: 115017.

[281]

Wang S T, Kravchyk K V, Pigeot-Rémy S, Tang W Q, Krumeich F, Wörle M, Bodnarchuk M I, Cassaignon S, Durupthy O, Zhao S L, Sanchez C, Kovalenko M V. ACS Appl. Nano Mater., 2019, 2(10): 6428.

[282]

Liu S N, Luo Z G, Tian G Y, Zhu M N, Cai Z Y, Pan A Q, Liang S Q. J. Power Sources, 2017, 363: 284.

[283]

Ni Q, Dong R Q, Bai Y, Wang Z H, Ren H X, Sean S, Wu F, Xu H J, Wu C. Energy Storage Mater., 2020, 25: 903.

[284]

Wang C, Zhang J S, Wang X F, Lin C F, Zhao X S. Adv. Funct. Mater., 2020, 30(45): 2002629.

[285]

Huo J H, Xue Y J, Cheng C, Guo S W. Ionics, 2020, 26(2): 971.

[286]

Jeong H, Noh Y, Lee D. Ceram. Int., 2019, 45(1): 985.

[287]

Raghu A V, Karuppanan K K, Nampoothiri J, Pullithadathil B. ACS Appl. Nano Mater., 2019, 2(3): 1152.

Funding

National Natural Science Foundation of China(52002076)

Natural Science Foundation of Shanghai(20ZR1403300)

Aerospace Science and Technology Innovation Fund(YF07050117F4051)

China Postdoctoral Science Foundation(2021M690644)

PDF (8709KB)

4991

Accesses

Citation

Detail

Sections

Recommended

Abstract

Graphical abstract

Keywords

Cite this article

1 Introduction

2 Crystal Structure and Lithium Storage Mechanism of TiO2;

2.1 Crystal structure

2.2 Two-phase solid solution mechanism of lithium storage in anatase TiO2:;

2.3 Intrinsic pseudocapacitance mechanism of lithium storage in TiO2(B):;

2.4 Potential-controlled phase transition of rutile TiO2:;

3 Modification Strategies of Anode Materials for TiO2;

3.1 Nanostructure dimension design

3.1.1 Zero-dimensional nanostructures

3.1.2 One-dimensional nanostructures

3.1.3 Two-dimensional nanostructures

3.1.4 Three-dimensional nanostructures

3.1.5 Multidimensional composite nanostructures

3.1.6 Micro-nano hierarchical structures

3.2 Intrinsic/extrinsic electronic structure control

3.2.1 Element doping

3.2.2 Ti3+ self-doping (oxygen vacancy)

3.2.3 Highly conductive material modified

3.3 Phase Structure Optimization

4 Performance of Li-ion battery based on TiO2 anode material

5 Conclusion and prospect

References

Funding

RIGHTS & PERMISSIONS

2 Crystal Structure and Lithium Storage Mechanism of TiO₂;

2.2 Two-phase solid solution mechanism of lithium storage in anatase TiO₂:;

2.3 Intrinsic pseudocapacitance mechanism of lithium storage in TiO₂(B):;

2.4 Potential-controlled phase transition of rutile TiO₂:;

3 Modification Strategies of Anode Materials for TiO₂;

3.2.2 Ti³⁺ self-doping (oxygen vacancy)

4 Performance of Li-ion battery based on TiO₂ anode material