Enhanced lithium storage property of Na-doped Li₂Na₂Ti₆O₁₄ anode materials for secondary lithium-ion batteries

Abstract

In this paper, a series of Na-doped Li₂Na₂Ti₆O₁₄ samples are synthesized by a simple solid-state reaction method through Li-site substitution with Na. Morphology observation shows that all five materials are well crystallized with a particle size in the range of 150–300 nm. Electrochemical analysis shows that Li_1.95Na_2.05Ti₆O₁₄ exhibits lower charge–discharge polarization (0.05 V) than that (0.11 V) of other Li_2−xNa_2+xTi₆O₁₄ samples (x = 0.00, 0.10, 0.15, 0.20). As a result, Li_1.95Na_2.05Ti₆O₁₄ has the highest initial charge capacity of 243.6 mA h g⁻¹, and maintains a reversible capacity of 210.7 mA h g⁻¹ after 79 cycles. For comparison, Li_2−xNa_2+xTi₆O₁₄ (x = 0.00, 0.10, 0.15 and 0.20) samples only hold a reversible capacity of 159.1, 203.5, 190.1 and 156.7 mA h g⁻¹, respectively. Moreover, Li_1.95Na_2.05Ti₆O₁₄ also delivers the best rate performance compared with the other four samples, with a charge capacity of 221.1 mA h g⁻¹ at 200 mA g⁻¹, 211.9 mA h g⁻¹ at 300 mA g⁻¹, and 198.7 mA h g⁻¹ at 400 mA g⁻¹. Besides, the reversible in situ structural evolution proves that Li_1.95Na_2.05Ti₆O₁₄ is a stable host for lithium storage. All the improved electrochemical properties of Na-doped Li₂Na₂Ti₆O₁₄ should be attributed to the Na-doping with low content, which reduces the charge–discharge polarization and improves the ionic conductivity.

---

1 Introduction

Because of their high energy density, high output working potential and long cycling life, rechargeable lithium ion batteries have been widely used in portable equipment, such as electric vehicles, hybrid electric vehicles and plug-in hybrid electric vehicles.^1,2 But the intrinsically poor safety characteristics hinder the large scale deployment of lithium ion batteries. One safety issue is that dendritic lithium may grow on the surface of the anode at high rates due to the low Li insertion potential of conventional graphite materials approaching 0.0 V versus Li/Li⁺.³

To replace the carbonous materials, various titanium based oxides have been proposed as high performance anode materials for lithium ion batteries.^4–9 Compared to widely reported Li₄Ti₅O₁₂, Li₂MTi₆O₁₄ (M = Sr, Ba, Pb) are a novel kind of hosts for lithium storage.^10–14 The stable framework, lower resistivity and lower working potential (1.4 V versus 1.55 V of Li₄Ti₅O₁₂) attract the materials scientists all over the world to make this compound become a promising lithium storage anode material. For instance, the solid state formation mechanism of Li₂SrTi₆O₁₄ anode is investigated using starting materials of SrCO₃, Li₂CO₃, and anatase TiO₂ by ex situ X-ray diffraction (XRD).¹⁰ It is found that the optimal calcination parameter to form pure Li₂SrTi₆O₁₄ is 950 °C for 8 h. To improve the packing density, mesoporous TiO₂ brookite is used as a template and reactant in a sol–gel synthesis.¹¹ As a result, Li₂SrTi₆O₁₄ anode shows impressive performance with a reversible capacity of 120 mA h g⁻¹ at C/14 rate and 92.0 mA h g⁻¹ at 4 C rate.

Isostructural to Li₂SrTi₆O₁₄, Li₂Na₂Ti₆O₁₄ reveals lower lithiation potential of 1.25 V with open channels in the structure enabling the reversible insertion of lithium ions.^15–19 Although Li₂Na₂Ti₆O₁₄ can be obtained at 600 °C by sol–gel method, it shows a low reversible capacity of about 75 mA h g⁻¹.¹⁵ To find appropriate preparation parameters, a comparison between sol–gel route and solid state method is undertaken for Li₂Na₂Ti₆O₁₄ at different sintering temperatures.¹⁶ It can be found that Li₂Na₂Ti₆O₁₄ prepared by sol–gel method at 700 °C reveals the highest initial charge specific capacity of 106.6 mA h g⁻¹ and the best rate properties than any other samples. To further improve the electrochemical properties, the coating by conductive additives, such as copper/carbon, carbon nanotube, graphene, carbon black, is made on Li₂Na₂Ti₆O₁₄.^17,18 As a result, modified Li₂Na₂Ti₆O₁₄ exhibits lower electrochemical polarization, quicker kinetic behavior and improved lithium storage capability compared to bare Li₂Na₂Ti₆O₁₄. However, few investigations have been reported to enhance the electrochemical performance of Li₂Na₂Ti₆O₁₄ by doping.²⁰

In this paper, five Na substituted Li_2−xNa_2+xTi₆O₁₄ (x = 0, 0.05, 0.10, 0.15, 0.20) samples are synthesized by a simple solid-state method. The structure, morphology and electrochemical properties of as-prepared sample are described and compared by using various analytical methods. An investigation is carried out to give new insights of Na doping into the evolutions of structural property, ion diffusion, rate performance, charge–discharge behavior of Li₂Na₂Ti₆O₁₄.

2 Experimental

2.1 Sample preparation

In the sample preparation procedure, all the chemical reagents are of analytical grade. Via a simple solid-state method, Li_2−xNa_2+xTi₆O₁₄ (x = 0, 0.05, 0.10, 0.15, 0.20) samples are prepared using stoichiometric amount of CH₃COOLi·2H₂O (99.0%), CH₃COONa·3H₂O (99%) and anatase TiO₂ (5–10 nm, 99.8%). The starting materials are mixed by planetary ball milling for 15 hours in ethanol to obtain homogeneous slurry. The resulting slurry is dried at 80 °C for 24 hours, and then calcined at 800 °C for 10 hours in air atmosphere to obtain the final Li_2−xNa_2+xTi₆O₁₄ samples.

2.2 Material characterization

The phase identifications are performed on a Bruker D8 Focus X-ray diffractometer with nickel-filtered Cu Kα radiation (λ = 1.5418 Å), operating at 40 kV and 40 mA. In situ X-ray diffraction (XRD) patterns are collected by the same instrument. The samples are scanned between 5 to 80° (2θ degree) using a scan speed of 0.2° min⁻¹. The surface morphologies of Li_2−xNa_2+xTi₆O₁₄ samples are observed by a L30 S-FEG field emission scanning electron microscope (SEM), conducting at 10 kV.

2.3 Electrode preparation and battery assembly

The working electrodes are prepared by dispersing a mixture of as-prepared active material, carbon black and polyvinylidene fluoride with a weight ratio of 8 : 1 : 1 in N-methyl-2-pyrrolidinone to form homogeneous slurry. After mixing, the slurry is pasted on copper foil evenly and then the film is dried at 120 °C in a vacuum oven for 12 hours, followed by cutting into sheets with a diameter of 15 mm. The two-electrode coin-type cells are assembled in an Ar-filled glove box by using Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.2) film as working electrode, lithium metal foil as counter electrode, Whatman glass fiber as separator and 1 mol L⁻¹ LiPF₆ in a 1 : 1 (v/v) mixture of ethylene carbonate and dimethyl carbonate as electrolyte.

2.4 Electrochemical evaluation

Galvanostatic charge–discharge tests are conducted on Land CT2001A multiple battery test system at a current density of 100 mA g⁻¹ between 0.0 and 3.0 V. Cyclic voltammograms (CVs) are performed via a computer-controlled CHI 660D electrochemical workstation at a scan rate of 0.1 mV s⁻¹ between 0.0 and 3.0 V. Electrochemical impedance spectroscopy (EIS) analysis is carried out by CHI 660D electrochemical workstation with an oscillating potential of 5 mV in the frequency ranging from 10⁵ to 10⁻² Hz. The cells for EIS observation are three-electrode system with metal lithium foils as reference and counter electrodes. All the tests are carried out at room temperature.

3 Results and discussion

Fig. 1 presents the XRD patterns of as-prepared Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) obtained at a temperature of 800 °C. All these patterns are collected on Bruker D8 Focus X-ray diffraction instrument at a room temperature. It can be observed from the XRD curves that the diffraction peak positions and the relative intensities of the prepared samples are well matched with the standard powder diffraction file of JCPDS card no. 52-0690, and it is clear that the diffraction patterns of all samples are similar with each other. Here, a twin peak appears at 65 °C in Fig. 1 corresponding to (824) and (040) peaks. It also shows from the graph that no impurities are observed in Li₂Na₂Ti₆O₁₄ and Na doped Li₂Na₂Ti₆O₁₄ samples. All the results obtained from the above means that the low content Na doping (0 ≤ x ≤ 0.2) does not change the main crystal structure of Li₂Na₂Ti₆O₁₄. For a clear observation, the peak position variation of (024) plane is enlarged and shown in Fig. 1b. The compounds with Na doping content of x = 0.05 and 0.10 exhibit no variation of the diffraction peak, revealing that low Na doping does not affect the lattice parameter of Li₂Na₂Ti₆O₁₄. However, the Bragg positions of the Na-doping Li_2−xNa_2+xTi₆O₁₄ samples with x = 0.15, 0.20 slightly shift to lower diffraction angles based on the enlarged (024) peak in Fig. 1b, indicating the lattice parameter of Li_2−xNa_2+xTi₆O₁₄ gradually increases after Na doping. The increased lattice parameter should be ascribed to the larger Na⁺ (0.97 Å) replacement of the smaller Li⁺ (0.68 Å) in the sites.

Fig. 1 XRD patterns of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples. (a) Full patterns, (b) local patterns.

Fig. 2 provides the typical SEM photographs of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples. It is apparent that all the five prepared samples are similar with each other, having a relative uniform morphology with narrow size distribution between 150 and 300 nm. It also can be found that the size of Li₂Na₂Ti₆O₁₄ is relatively smaller than another four samples, which may attribute to the bigger Na⁺ replacement of the smaller Li⁺ in the structure.

Fig. 2 SEM images of Li_2−xNa_2+xTi₆O₁₄ samples. (a and b) x = 0.00, (c and d) x = 0.05, (e and f) x = 0.10, (g and h) x = 0.15, (i and j) x = 0.20.

The CV curves of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) are presented in Fig. 3. All the tests are conducted at room temperature under a scan rate of 0.1 mV s⁻¹ between 0.0 and 3.0 V. As is shown in Fig. 3a–e, a pair of characteristic redox peaks can be observed at around 1.17 and 1.19 V for five samples, regarded as the signature of two lithium-ion per formula insertion into and extraction from the Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) framework,^14,16 which is in consistence with the charge–discharge potential plateaus in Fig. 4. According to the formula of Li_2−xNa_2+xTi₆O₁₄, a total of six Ti⁴⁺ can be reduced to Ti³⁺ after a theoretical evaluation. Thus, the broad peak around 0.0–0.1 V is associated with another four lithium-ion formula storage in the compound. And all the CV peaks are narrow and sharp, corresponding to the rapid kinetic process for lithium ions transportation in the lattices. Besides, pristine Li₂Na₂Ti₆O₁₄ and Li_1.95Na_2.05Ti₆O₁₄ exhibit much obvious reduction peaks at 1.178 V and 1.205, respectively, and a sharper oxidation peaks at 1.362 V than other Li_2−xNa_2+xTi₆O₁₄ (0.10 ≤ x ≤ 0.20) samples. The peak current reduces relatively rapid along with the increase of Na doping concentration from 0.05 to 0.20, revealing a drop in the electrode kinetics. Moreover, the initial three CV curves of Li₂Na₂Ti₆O₁₄ and Li_1.95Na_2.05Ti₆O₁₄ coincide with each other and show higher peak current than the other three ones, indicates that Li₂Na₂Ti₆O₁₄ and Li_1.95Na_2.05Ti₆O₁₄ have higher electrochemical reaction activity and reversibility. The difference between anodic and cathodic peaks can be mainly attributed to the slow lithium ion diffusivity in solid-state body of bulk Li_2−xNa_2+xTi₆O₁₄ (0.10 ≤ x ≤ 0.20).¹³ The potential difference (Δφ_p) of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) electrodes between oxidation and reduction peaks is listed in Table 1. It is obvious that the potential separation (Δφ_p value) of pristine Li₂Na₂Ti₆O₁₄ (about 1.82 mV) is much larger than those of Na-doped samples. It suggests that Na doping is beneficial to the reversible intercalation and deintercalation of Li⁺ in the structure, and then enhances the reversibility of Li₂Na₂Ti₆O₁₄. Furthermore, Li_1.95Na_2.05Ti₆O₁₄ expresses the smallest potential separation among all the materials, indicating that Li_1.95Na_2.05Ti₆O₁₄ may own the best electrochemical performance for its fast electron transfer kinetics and outstanding cycling reversibility.

Fig. 3 CVs of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples. (a) Li₂Na₂Ti₆O₁₄, (b) Li_1.95Na_2.05Ti₆O₁₄, (c) Li_1.9Na_2.1Ti₆O₁₄, (d) Li_1.85Na_2.15Ti₆O₁₄, (e) Li_1.8Na_2.2Ti₆O₁₄.

Fig. 4 The 1^st (a), 10^th (b), 20^th, (c) 50^th (d) and 79^th (e) charge–discharge curves of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples at a current density of 100 mA g⁻¹.

Table 1 Potentials of the redox peaks in the CVs for Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples

Samples	φpa (V)	φpc (V)	Δφp^a (mV)
a Δφp = φpa − φpc.
Li2Na2Ti6O14	1.360	1.178	182
Li1.95Na2.05Ti6O14	1.362	1.205	157
Li1.9Na2.1Ti6O14	1.355	1.190	165
Li1.85Na2.15Ti6O14	1.362	1.198	164
Li1.8Na2.2Ti6O14	1.354	1.186	168

---

Fig. 4 depicts the charge and discharge curves upon the 1^st, 10^th, 20^th and 50^th cycles of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples at the potential range of 0.0–3.0 V. As is shown in Fig. 4a, all above-mentioned compounds exhibit one charge plateau and one discharge plateau, similar to the previous report of Li₂Na₂Ti₆O₁₄.^14–17 For the initial cycle of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20), the potential drops rapidly from open-circuit potential to 1.35 V, and then the discharge potential maintains for a long flat platform at a average potential of 1.24 V. Finally, a long slope appears between 0.0 and 1.2 V. During the charging process, the charge plateaus of Li_2−xNa_2+xTi₆O₁₄ (x = 0.00, 0.10, 0.15 and 0.20) are at about 1.34 V. In contrast, the plateau of Li_1.95Na_2.05Ti₆O₁₄ keeps nearly at 1.30 V, and then rapidly increases to the cut-off potential. It reveals that Li_1.95Na_2.05Ti₆O₁₄ has lower charge–discharge polarization (0.05 V) than that (0.11 V) of Li_2−xNa_2+xTi₆O₁₄ (x = 0.00, 0.10, 0.15, 0.20), which is also in correspondence with the results of CV curves. Besides, it is evident that the charge plateau of Li_1.95Na_2.05Ti₆O₁₄ are much longer than those of Li_2−xNa_2+xTi₆O₁₄ (x = 0.00, 0.10, 0.15, 0.20) in all of four graphs. After 50 cycles, Li_1.95Na_2.05Ti₆O₁₄ still exhibits the discharge and charge plateaus at around 1.16 and 1.33 V, respectively. Its charge–discharge polarization remains much lower than those of Li_2−xNa_2+xTi₆O₁₄ (x = 0.00, 0.10, 0.15, 0.20), which further confirm the superior cycling performance of Li_1.95Na_2.05Ti₆O₁₄.

To further conduct investigation of the electrochemical properties for Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20), the cycling performance and corresponding cycling Coulombic efficiencies upon repeated cycles are also shown in Fig. 5. It can be clearly observed that Li_1.95Na_2.05Ti₆O₁₄ express the highest charge capacity of 243.6 mA h g⁻¹ for the first cycle, while Li₂Na₂Ti₆O₁₄, Li_1.9Na_2.1Ti₆O₁₄, Li_1.85Na_2.15Ti₆O₁₄ and Li_1.8Na_2.2Ti₆O₁₄ only deliver the initial charge capacity of 240.1, 229.0, 214.0 and 225.2 mA h g⁻¹, respectively. Besides, Li_1.95Na_2.05Ti₆O₁₄ still remains the charge capacity of 210.7 mA h g⁻¹ after 79 repeated cycles, which proves that appropriate amount Na doping is beneficial to improve the cycling stability of Li₂Na₂Ti₆O₁₄. On the other hand, Li₂Na₂Ti₆O₁₄, Li_1.9Na_2.1Ti₆O₁₄, Li_1.85Na_2.15Ti₆O₁₄ and Li_1.8Na_2.2Ti₆O₁₄ only hold the reversible capacity of 159.1, 203.5, 190.1 and 156.7 mA h g⁻¹, respectively, implying high Na doping content goes against the cycling performance. Therefore, it is known that the superior cycling performance of Li_1.95Na_2.05Ti₆O₁₄ is attributed to the low content Na doping in the structure, which may improve the lithium ion diffusivity and structural stability of Li₂Na₂Ti₆O₁₄. Furthermore, the corresponding cycling Coulombic efficiencies upon repeated cycles are shown in Fig. 5b. Among the five compounds, Li_1.95Na_2.05Ti₆O₁₄ still presents the highest average cycling efficiency of 99.0%, while the others hold the cycling Coulombic efficiency of nearly 94.5–98.0%, which further indicates the outstanding cycling performance of Li_1.95Na_2.05Ti₆O₁₄.

Fig. 5 (a) Cycle performance and (b) corresponding Coulombic efficiency of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples.

Fig. 6a displays the Nyquist plots before cycles for the as-prepared Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20). The enlarged Nyquist curves of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) in the high frequency region are also shown in Fig. 6a as an insert graph. All the curves have one semi-circle in the high-frequency region and a straight line in low-frequency region, which are related with the charge transfer process and lithium ion diffusion behavior. The EIS patterns can be used to calculate the lithium diffusion coefficient (D_Li), which is an important factor to evaluate the electrochemical kinetics of compounds. Here, the D_Li of lithium ion can be calculated by the following equation:^4,21

where R is the gas constant (8.314 J mol⁻¹ K⁻¹), T is the absolute temperature for battery testing (298 K), A is the surface area of electrode (1.766 cm²), n is the number of electrons transfer during the electrochemical reaction, F is the Faraday constant (94 850 C mol⁻¹), C_Li is the concentration of lithium ion in the compound, and the Warburg factor σ can be calculated from the plots in the low-frequency region, and the relationship of σ with Z_re is as follows:

Fig. 6 (a) EIS patterns before cycles and (b) corresponding Z_re vs. ω^−0.5 curves at the low frequency range for Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples.

The Z_re–σ plots are shown in Fig. 6b, and a linear characteristic could be seen from the curves. Based on the eqn (1) and (2), the calculated diffusion coefficients are listed in Table 2. Observed from Table 2, it is clear that Li_1.95Na_2.05Ti₆O₁₄ possesses the highest lithium ion diffusion coefficient of 1.11 × 10⁻¹⁴ cm² s⁻¹, while the pristine Li₂Na₂Ti₆O₁₄ only reveals a low lithium ion diffusion coefficient of 1.57 × 10⁻¹⁵ cm² s⁻¹. With the increase of Na doping content, the lithium ion diffusion coefficient finally reduces to 4.37 × 10⁻¹⁵ cm² s⁻¹ with x = 2.0 in Li_2−xNa_2+xTi₆O₁₄. It suggests that appropriate amount Na doping is beneficial to improve the lithium ion diffusivity and kinetic behavior of Li₂Na₂Ti₆O₁₄. Therefore, it is expected that Li_1.95Na_2.05Ti₆O₁₄ may have an outstanding rate property.

Table 2 The Li⁺ diffusion coefficients calculated from EIS patterns for Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples

Samples	x = 0.00	x = 0.05	x = 0.10	x = 0.15	x = 0.20
DLi (cm^2 s^−1)	1.57 × 10^−15	1.11 × 10^−14	1.13 × 10^−14	2.81 × 10^−15	4.37 × 10^−15
σ (Ω s^−0.5)	615.5	226.5	229.2	460.1	368.6

---

Fig. 7 delivers the initial charge–discharge curves of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples at 200, 300 and 400 mA g⁻¹. It can be found that Li₂Na₂Ti₆O₁₄, Li_1.95Na_2.05Ti₆O₁₄ and Li_1.9Na_2.1Ti₆O₁₄ show a flat plateau at around 1.34 V for the charge plot and a flat plateau at about 1.20 V for the discharge plot. In contrast, Li_1.85Na_2.15Ti₆O₁₄ and Li_1.8Na_2.2Ti₆O₁₄ display the discharge and charge plateaus at 0.98 and 1.48 V, respectively. The results indicate that Li₂Na₂Ti₆O₁₄, Li_1.95Na_2.05Ti₆O₁₄ and Li_1.9Na_2.1Ti₆O₁₄ maintain lower electrochemical polarization than Li_1.85Na_2.15Ti₆O₁₄ and Li_1.8Na_2.2Ti₆O₁₄ during high rate discharge–charge cycles, even at a current density of 400 mA g⁻¹. It also suggests that Li₂Na₂Ti₆O₁₄, Li_1.95Na_2.05Ti₆O₁₄ and Li_1.9Na_2.1Ti₆O₁₄ may have better kinetic properties at high rates.

Fig. 7 The comparative charge–discharge curves of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples at different current densities. (a) 200 mA g⁻¹, (b) 300 mA g⁻¹, and (c) 400 mA g⁻¹.

The rate performances of five Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples are presented in Fig. 8. Extended galvanostatic cycles of the Li_2−xNa_2+xTi₆O₁₄ electrodes are performed at increasing current density of 150, 200, 250, 300, 350 and 400 mA g⁻¹. Viewed from Fig. 8, it is obvious that the pristine Li₂Na₂Ti₆O₁₄ only maintains the lithium storage capacity of 206.5 mA h g⁻¹ at 200 mA g⁻¹, 194.6 mA h g⁻¹ at 300 mA g⁻¹, and 186.6 mA h g⁻¹ at 400 mA g⁻¹. With the increase of Na doping content, the rate performance of Li_2−xNa_2+xTi₆O₁₄ shows an improvement at low Na content and then presents a decrease at high Na content. Compared with other four samples, Li_1.95Na_2.05Ti₆O₁₄ apparently delivers a better performance with the charge capacity of 221.1 mA h g⁻¹ at 200 mA g⁻¹, 211.9 mA h g⁻¹ at 300 mA g⁻¹, and 198.7 mA h g⁻¹ at 400 mA g⁻¹. This result is roughly consistent with the cycling performance as shown in Fig. 5a. From what have discussed above, it can be concluded that the incorporation of low sodium content x = 0.05 provokes positive effect on ionic conductivity and the electrochemical properties for Li₂Na₂Ti₆O₁₄.

Fig. 8 Rate performance of Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples.

To observe the structural evolution of Na-doped Li₂Na₂Ti₆O₁₄, in situ XRD technique is used to investigate the reversibility of Li_1.95Na_2.05Ti₆O₁₄. The description of in situ cell and its preparation process can be found in our previous paper.²² Fig. 9–11 shows that the characteristic diffraction peaks of Li_1.95Na_2.05Ti₆O₁₄ gradually shift to lower Bragg positions during the lithiation process. With a reverse delithiation reaction, all the featured peaks can go back the original positions. It suggests that the structural change of Li_1.95Na_2.05Ti₆O₁₄ is highly reversible during the charge–discharge process. As a result, the stable host structure of Li_1.95Na_2.05Ti₆O₁₄ can ensure the long-term repeated electrochemical cycles.

Fig. 9 Overall in situ XRD patterns of Li_1.95Na_2.05Ti₆O₁₄ during the charge–discharge process.

Fig. 10 Selected in situ XRD patterns of Li_1.95Na_2.05Ti₆O₁₄ with the same background.

Fig. 11 The evolution of relative intensity versus Bragg position of Li_1.95Na_2.05Ti₆O₁₄ during the charge–discharge process.

4 Conclusions

Via Li-site substitution with Na, five Li_2−xNa_2+xTi₆O₁₄ (0 ≤ x ≤ 0.20) samples are prepared by a simple solid-state method. XRD results reveal that all the as-prepared samples are single-phase without any impurity. Morphological characterization shows that all the five samples are well-crystallized products with a homogeneous particle size between 150 and 300 nm. Electrochemical analysis shows that Li_1.95Na_2.05Ti₆O₁₄ exhibits the highest reversible capacity and the best rate property among all the Li₂Na₂Ti₆O₁₄-type anodes. Compared with other four samples, Li_1.95Na_2.05Ti₆O₁₄ can deliver a better rate performance with the charge capacity of 221.1 mA h g⁻¹ at 200 mA g⁻¹, 211.9 mA h g⁻¹ at 300 mA g⁻¹, and 198.7 mA h g⁻¹ at 400 mA g⁻¹. EIS results prove that the improved lithium storage performance of Na-doped Li₂Na₂Ti₆O₁₄ is related to the improved ionic conductivity and the decreased redox polarization via Li-site substitution with Na. Besides, the reversible structural change of Li_1.95Na_2.05Ti₆O₁₄ as observed by in situ XRD proves its stable host structure for repeated lithium storage.

Acknowledgements

This work is sponsored by Ningbo Key Innovation Team (2014B81005) and Ningbo Natural Science Foundation (2014A610042). The work is also supported by opening project of Key Laboratory of Photochemical Conversion and Optoelectronic Materials, TIPC, CAS (PCOM201408) and K. C. Wong Magna Fund in Ningbo University.

References

1. J. M. Tarascon and M. Armand, Issues and challenges facing rechargeable lithium batteries, Nature, 2001, 414, 359–367.
2. M. Armand and J. M. Tarascon, Building better batteries, Nature, 2008, 451, 652–657.
3. S. S. Zhang, The effect of the charging protocol on the cycle life of a Li-ion battery, J. Power Sources, 2006, 161, 1385–1391.
4. T. Yi, H. Liu, Y. Zhu, L. Jiang, Y. Xie and R. Zhu, Improving the high rate performance of Li₄Ti₅O₁₂ through divalent zinc substitution, J. Power Sources, 2012, 215, 258–265.
5. T. Takashima, T. Tojo, R. Inada and Y. Sakurai, Characterization of mixed titanium–niobium oxide Ti₂Nb₁₀O₂₉ annealed in vacuum as anode material for lithium-ion battery, J. Power Sources, 2015, 276, 113–119.
6. J. Akimoto, K. Kataoka, N. Kojima, S. Hayashi, Y. Gotoh, T. Sotokawa and Y. Kumashiro, A novel soft-chemical synthetic route using Na₂Ti₆O₁₃ as a starting compound and electrochemical properties of H₂Ti₁₂O₂₅, J. Power Sources, 2013, 244, 679–683.
7. W. Cho, M. S. Park, J. H. Kim and Y. J. Kim, Interfacial reaction between electrode and electrolyte for a ramsdellite type Li_2+xTi₃O₇ anode material during lithium insertion, Electrochim. Acta, 2012, 63, 263–268.
8. J. C. Pérez-Flores, A. Kuhn and F. García-Alvarado, Synthesis, structure and electrochemical Li insertion behaviour of Li₂Ti₆O₁₃ with the Na₂Ti₆O₁₃ tunnel-structure, J. Power Sources, 2011, 196, 1378–1385.
9. J. Liu, Y. Li, X. Wang, Y. Gao, N. Wu and B. Wu, Synthesis process investigation and electrochemical performance characterization of SrLi₂Ti₆O₁₄ by ex situ XRD, J. Alloys Compd., 2013, 581, 236–240.
10. D. Dambournet, I. Belharouak, J. Ma and K. Amine, Template-assisted synthesis of high packing density SrLi₂Ti₆O₁₄ for use as anode in 2.7 V lithium-ion battery, J. Power Sources, 2011, 196, 2871–2874.
11. J. Liu, X. Sun, Y. Li, X. Wang, Y. Gao, K. Wu, N. Wu and B. Wu, Electrochemical performance of LiCoO₂/SrLi₂Ti₆O₁₄ batteries for high-power applications, J. Power Sources, 2014, 245, 371–376.
12. I. Koseva, J. P. Chaminade, P. Gravereau, S. Pechev, P. Peshev and J. Etourneau, A new family of isostructural titanates, MLi₂Ti₆O₁₄ (M = Sr, Ba, Pb), J. Alloys Compd., 2005, 389, 47–54.
13. X. T. Lin, P. Li, L. Y. Shao, M. Shui, D. J. Wang, N. B. Long, Y. L. Ren and J. Shu, Lithium barium titanate: a stable lithium storage material for lithium-ion batteries, J. Power Sources, 2015, 278, 546–554.
14. D. Dambournet, I. Belharouak and K. Amine, MLi₂Ti₆O₁₄ (M = Sr, Ba, 2Na) lithium insertion titanate materials: a comparative study, Inorg. Chem., 2010, 49, 2822–2826.
15. S. Y. Yin, L. Song, X. Y. Wang, Y. H. Huang, K. L. Zhang and Y. X. Zhang, Reversible lithium storage in Na₂Li₂Ti₆O₁₄ as anode for lithium ion batteries, Electrochem. Commun., 2009, 11, 1251–1254.
16. K. Wu, D. Wang, X. Lin, L. Shao, M. Shui, X. Jiang, N. Long, Y. Ren and J. Shu, Comparative study of Na₂Li₂Ti₆O₁₄ prepared by different methods as advanced anode material for lithium-ion batteries, J. Electroanal. Chem., 2014, 717–718, 10–16.
17. K. Wu, J. Shu, X. Lin, L. Shao, M. Lao, M. Shui, P. Li, N. Long and D. Wang, Enhanced electrochemical performance of sodium lithium titanate by coating various carbons, J. Power Sources, 2014, 272, 283–290.
18. K. Wu, X. Lin, L. Shao, M. Shui, N. Long, Y. Ren and J. Shu, Copper/carbon coated lithium sodium titanate as advanced anode material for lithium-ion batteries, J. Power Sources, 2014, 259, 177–182.
19. K. Wu, J. Shu, X. Lin, L. Shao, P. Li, M. Shui, M. Lao, N. Long and D. Wang, Phase composition and electrochemical performance of sodium lithium titanates as anode materials for lithium rechargeable batteries, J. Power Sources, 2015, 275, 419–428.
20. M. Lao, X. Lin, P. Li, L. Shao, K. Wu, M. Shui, N. Long, Y. Ren and J. Shu, Preparation and electrochemical characterization of Li_2+xNa_2−xTi₆O₁₄ (0≤ x ≤0.2) as anode materials for lithium-ion batteries, Ceram. Int., 2015, 41, 2900–2907.
21. T. Yi, S. Yang, X. Li, J. Yao, Y. Zhu and R. Zhu, Sub-micrometric Li_4−xNa_xTi₅O₁₂ spinel as anode material exhibiting high rate capability, J. Power Sources, 2014, 246, 505–511.
22. J. Shu, M. Shui, D. Xu, Y. L. Ren, D. J. Wang, Q. C. Wang, R. Ma, W. D. Zheng, S. Gao, L. Hou, J. J. Xu, J. Cui, Z. H. Zhu and M. Li, Large-scale synthesis of Li_1.15V₃O₈ nanobelts and their lithium storage behavior studied by in situ X-ray diffraction, J. Mater. Chem., 2012, 22, 3035–3043.

---