Introduction

The using of Li4Ti5O12 as anode for lithium-ion batteries was first proposed by Canadian scholar K. Zaghib [1,2,3]. Later, some people also used Li4Ti5O12 as anode material [4,5,6,7]. It is found that [8,9,10,11] the structure of Li4Ti5O12 does not change during charging and discharging. The Li4Ti5O12 material has the advantages of high charge-discharge efficiency, where the diffusion coefficient of Li+ is ten times higher than carbon anode [12,13,14,15]. It has faster charging characteristics, more cycles, and safer use of the new-generation lithium-ion battery [16,17,18,19,20].

The low migration ability of Li+ in the material is a bottleneck that inhibits development of LiFePO4 cathode materials [21,22,23,24]. At present, the mature methods such as carbon coating, particle size reduction, and metal element doping are used to improve the low electronic conductivity of LiFePO4 and the small diffusion coefficient of lithium ions [25,26,27,28,29]. Studies have shown that [30,31,32,33,34] doping high-valence Ti4+ in LiFePO4 can cause VLi defects in the material, thereby improving Li+ migration ability and electrochemical performance of the material. Li4Ti5O12 and LiFePO4 materials were selected as research objects, which were prepared by solid-phase method. The two materials were simply blended in different proportions to obtain LiFePO4/C-Li4Ti5O12 composites. The structure, morphology, and electrochemical properties of LiFePO4/C-Li4Ti5O12 composite cathode materials were investigated.

Experiment

Preparation of LiFePO4/C

LiOH·H2O and FePO4 with a molar ratio of 1.05:1 were mixed as raw materials, and ethanol was used as a dispersing agent. The ball-to-material ratio was 15:1 in a planetary ball mill. After ball milling for 10 h, it was dried in a vacuum-drying oven. The dried mixture was pre-sintered at 350 °C for 4 h under N2 atmosphere and then calcined at 700 °C for 12 h to obtain LiFePO4/C material.

Preparation of Li4Ti5O12

LiOH·H2O and TiO2 with a molar ratio of 4.08:5 were mixed as raw materials, ethanol was used as dispersing agent, and the ball-to-material ratio was 15:1 in a planetary ball mill. The ball was ground for 10 h, and the ground powder was placed in a vacuum-drying oven. The dried mixture was pre-sintered at 450 °C for 4 h under N2 atmosphere and then calcined at 800 °C for 12 h to obtain Li4Ti5O12 material.

Preparation of LiFePO4/C-Li4Ti5O12 composite

The LiFePO4/C and Li4Ti5O12 materials were placed in a planetary ball mill with ethanol as dispersing agent. The ratio of the ball to the material was 15:1. After ball milling for 10 h, the material was dried and transferred into a porcelain boat. Under nitrogen protection, the mixture was pre-sintered at 300 °C for 5 h and then calcined at 500 °C for 14 h to obtain a sample of LiFePO4/C-Li4Ti5O12. LiFePO4/C-Li4Ti5O12 composite cathode materials with different composite ratios were synthesized from LiFePO4/C and Li4Ti5O12 at a mass ratio of 9:1, 8:2, 7:3, and 6:4, respectively, which were identified as LFP-9, LFP-8, LFP-7, and LFP-6.

Assembly of button battery

The active material, polyvinylidene fluoride and acetylene black were taken at a mass ratio of 80:10:10, and NMP was mixed uniformly to adjust the slurry. Finally, the slurry was coated on an aluminum foil and dried in a vacuum oven at 80 °C for 20 h. After the sample was dried, it was punched into a disk having a diameter of 17.2 mm by a punching machine as a positive electrode sheet. The lithium sheet was a negative electrode, the Celgard 2400 porous polypropylene was used as separator, and the electrolyte was an organic solution of EC, DMC, and DEC (volume ratio EC:DMC:DEC = 1:1:1) of 1 M LiPF6. A button battery was assembled in the glove box for testing.

Results and discussion

Structure analysis of materials

Figure 1 shows the XRD patterns of LiFePO4/C and Li4Ti5O12, which are consistent with the standard spectrum. There is no impurity phase in diffraction peaks, indicating that the prepared LiFePO4/C sample is a pure-phase olivine structure. After the carbon source is added, the carbon exists in the material at amorphous state. The Li4Ti5O12 material is a spinel structure. Compared with the XRD of LiFePO4/C material, each diffraction peak of Li4Ti5O12 material has higher intensity, sharper peak shape, and narrower half width, which indicate that the Li4Ti5O12 sample has better crystallinity than LiFePO4/C material.

Fig. 1
figure 1

The XRD patterns of LiFePO4 and Li4Ti5O12 samples

Full size image

Figure 2 shows that the crystal structures of the two materials are not changed after the simple blending of LiFePO4/C and Li4Ti5O12. The diffraction peaks are superposed on each other according to the blending ratio, and no impurity peak appears. With the gradual increase of the proportion of Li4Ti5O12 in the composite, the characteristic diffraction peak of Li4Ti5O12 is increasingly obvious.

Fig. 2
figure 2

The XRD patterns of LFP-6, LFP-7, LFP-8, and LFP-9 samples

Full size image

Morphological analysis of materials

As shown in Fig. 3(A, a), LiFePO4/C is a secondary particle with an average particle size of about 18 μm. Each secondary particle is composed of many primary particles, and the LiFePO4/C sample reflected by the primary particles has an olivine structure. It can be seen from Fig. 3(B, b) that Li4Ti5O12 material is composed of primary particles of spindle type, and the average particle diameter is 7 μm. These particles are not agglomerated into secondary particles, and the dispersibility is well. The Li4Ti5O12 sample has small particle size and large surface area, which facilitates the deintercalation of Li+.

Fig. 3
figure 3

SEM of LiFePO4/C (A, a) and Li4Ti5O12 (B, b) samples

Full size image

As illustrated in Fig. 4, the morphology of the LiFePO4/C-Li4Ti5O12 composite obtained after blending is not as good as that of the single LiFePO4/C, which may be due to the long-time grinding of the blended material. After that, the morphology of the olivine particles of LiFePO4/C material is destroyed. The smaller particles of Li4Ti5O12 are all adhered to the surface of LiFePO4/C particles and a part of the encapsulated LiFePO4/C particles agglomerate. From Fig. 4(A, a), we find that LiFePO4/C particles are not completely covered by Li4Ti5O12 appeared in the LFP-9 material. This is because the proportion of Li4Ti5O12 in the blended material is low and the effect of complete coating cannot be achieved. This will affect the electrochemical performance to some extent. We can also know that the particles in LFP-8 material are smaller than the particles of LFP-9, LFP-7, and LFP-6 materials and the particle size is average. Li4Ti5O12 cannot only completely coat LiFePO4/C but be dispersed in LiFePO4/C particle gap with a tight filling state, which is conducive to the transfer of lithium ions. It has a positive effect on the improvement of electrochemical performance.

Fig. 4
figure 4

SEM of LFP-9 (A, a), LFP-8 (B, b), LFP-7 (C, c), and LFP-6 (D, d) samples

Full size image

Figure 5 shows that the four elements of Ti, P, O, C, and Fe are uniformly distributed in the samples. Combined with the SEM and XRD patterns of LFP-8, the synthesized material is a pure LiFePO4/C-Li4Ti5O12 with uniform element distribution.

Fig. 5
figure 5

The EDS diagrams of synthesized material LFP-8

Full size image

Electrochemical performance analysis of materials

Charge and discharge test of materials

Figure 6 shows LiFePO4/C samples tested for charge and discharge at a voltage range of 2.0–4.8 V at 0.1, 0.2, 0.5, 1, and 2C, respectively. As can be seen from Fig. 6, the LiFePO4/C sample exhibits a charge and discharge characteristic platform at about 3.6 V/3.4 V, which corresponds to the redox reaction of Fe3+/Fe2+. The specific discharge capacities at 0.1, 0.2, 0.5, 1, and 2C rates are 128, 121, 108, 72, and 39 mAh g−1, respectively. The corresponding coulombic efficiencies are 98.1, 96.7, 96.6, 91.4, and 84.4%, respectively. It can be seen from the data that as the charge-discharge rate increases, the discharge-specific capacity and discharge efficiency of the LiFePO4/C sample decrease rapidly, and the charge-discharge platform gradually becomes unapparent. The potential difference between the platforms and the polarization of the battery increases, the discharge efficiency decreases, resulting in the worse rate performance of the battery.

Fig. 6
figure 6

The first charge-discharge capacity curves of LiFePO4/C samples under different ratios

Full size image

Figure 7 shows the first charge-discharge curves of LFP-9, LFP-8, LFP-7, and LFP-6 composites at a voltage range of 1.0–4.2 V at 0.1C. It can be seen from Fig. 7 that the charge-discharge curve of LiFePO4/C-Li4Ti5O12 composite combines the characteristics of charge and discharge curves of LiFePO4/C and Li4Ti5O12 materials. During the charging process, a platform appears at about 1.5 V, corresponding to the lithium intercalation process of Li4Ti5O12 material, and another platform appears at 3.5 V, corresponding to the lithium intercalation process of LiFePO4/C material. Subsequent discharges exhibit similar voltage curves, indicating that the material has reversible structural changes during lithium ion insertion and removal. At 0.1C rate, the initial discharge-specific capacities of LFP-9, LFP-8, LFP-7, and LFP-6 composites are 147, 152, 149, and 142 mAh g−1. Comparing LiFePO4/C samples, the specific discharge capacity of the composite is higher than that of the single LiFePO4/C material, because Li4Ti5O12 is also involved in the electrochemical reaction in the composite. Since the two materials have the first irreversible capacity loss, the coulomb effect of the Li4Ti5O12 material is higher than that of LiFePO4/C at the same rate. Therefore, the first coulomb efficiency of the composite electrode is higher than that of the single LiFePO4/C material. Among them, LFP-8 material exhibits the highest discharge-specific capacity and superior discharge efficiency.

Fig. 7
figure 7

The first charge-discharge capacity of LFP-9, LFP-8, LFP-7, and LFP-6 composite samples

Full size image

Cycle performance test

Figure 8 is a cycle curve of single LiFePO4/C, LFP-9, LFP-8, LFP-7, and LFP-6 composites cycled at 0.1C for 100 times. After cycles, the capacity retention rates of single LiFePO4/C and composite materials LFP-9, LFP-8, LFP-7, and LFP-6 are 89.3, 96.5, 97.3, 96.5, and 96.1%, respectively. In comparison, the LiFePO4/C sample has poor cycle performance. The discharge capacity decay rate reaches 10.7% after 100 charge and discharge cycles. As can be seen from the curve, the cycle performance of LiFePO4/C-Li4Ti5O12 composite is better than that of single LiFePO4/C material. This is because Li4Ti5O12 has better cycling performance which plays a vital role in improving the cycle performance of LiFePO4/C-Li4Ti5O12 composite. When the recombination ratio is 8:2, the capacity retention ratio is 97.3%.

Fig. 8
figure 8

The cycle curves of LiFePO4/C, Li4Ti5O12, LFP-9, LFP-8, LFP-7, and LFP-6 materials under 0.1C ratio

Full size image

Cyclic voltammetry

Figure 9 shows the cyclic voltammetry of LiFePO4/C and Li4Ti5O12 materials measured at a scan speed of 0.1 mV s−1. The corresponding scan voltage ranges are 2.0–4.6 and 1.0–2.3 V, respectively. We can see that two pairs of symmetry oxidation/reduction peak appear on the curve, which correspond to the deintercalation and intercalation process of Li+. Figure 9 shows the potential difference between the oxidation potential and the reduction potential of LiFePO4 is 0.42 V, indicating there is a certain polarization phenomenon during the electrode reaction. It is observed that the area of the oxidation peak is larger than the area of the reduction peak. In the process of charge and discharge, the discharge-specific capacity is larger than the charge-specific capacity, and on the cyclic voltammogram, the area of the reduction peak is larger than that of the oxidation peak. The equilibrium potential of Ti4+/Ti3+ in Li4Ti5O12 is 1.55 V. As we can see in the figure, the peak potential of the oxidation peak during charging is greater than the equilibrium potential, while the peak potential of the reduction peak during the discharge is less than equilibrium. The potential difference between the oxidation potential and the reduction potential is 0.35 V, which is significantly smaller than the LiFePO4 potential difference, indicating the polarization of the Li4Ti5O12 material is smaller than that of LiFePO4. It is observed that the oxidation peak area and the reduced peak area of Li4Ti5O12 are basically the same, indicating the irreversible capacity loss of this material is less in the charging and discharging process.

Fig. 9
figure 9

The CV curves of LiFePO4/C and Li4Ti5O12 materials

Full size image

Figure 10 shows the cyclic voltammogram of LFP-9, LFP-8, LFP-7, and LFP-6 composite samples at room temperature with a scan speed of 0.1 mV s−1. The voltage is between 1.0 and 4.2 V. As can be seen from the figure, there are two pairs of redox peak pairs appearing on the curve. The redox peak pair at the low voltage corresponds to intercalation and deintercalation of Li+ in Li4Ti5O12 material. The redox peak pair at the high voltage is derived from the redox reaction of Fe2+/Fe3+ in LiFePO4 material. The peak potential difference between the redox peaks reflects the degree of reversibility of the electrochemical, and the peak current density reflects the electrochemical reaction rate. When the blending ratio is 9:1 (LFP-9), the symmetry between the oxidation peak and the reduction peak is the best. The potential difference is the smallest. This shows that the polarization of the battery is minimum under the blend ratio while the discharge efficiency of the battery and the reversibility of the battery is optimal. The charge and discharge amount of the battery is the largest when the blending ratio is 8:2 (LFP-8). This is consistent with the conclusions from the previous charge and discharge test results.

Fig. 10
figure 10

The CV curves of LFP-9, LFP-8, LFP-7, and LFP-6 composite materials under different composite ratios

Full size image

Electrochemical impedance spectroscopy analysis

Figure 11 shows the EIS and equivalent circuit of LiFePO4/C, Li4Ti5O12, LFP-9, LFP-8, LFP-7, and LFP-6 materials. The frequency range of the test is 1 × 10−2–1 × 105 Hz. The diameter of the semicircle reflects the charge transfer resistance (Rct) of the electrode. The semicircle of the high-frequency region to the intermediate-frequency region reflects the internal resistance (RSEI) caused by the solid electrolyte membrane on the electrode surface. RW stands for Warburg resistance, RS is the resistance of the electrolyte and other ohmic contact resistance in the cell. CPE1 and CPE2 are the capacitors corresponding to RSEI and Rct.

Fig. 11
figure 11

The EIS curves and equivalent circuit of LiFePO4/C, Li4Ti5O12, LFP-9, LFP-8, LFP-7, and LFP-6 materials

Full size image

As can be seen from the figure, the impedance of Li4Ti5O12 material is much smaller than that of other materials, while the impedance of LiFePO4/C material is much larger than that of other materials. It is shown that Li4Ti5O12 material is easier to overcome the dynamic charge binding during charge and discharge, which is more conducive to Li+ intercalation and deintercalation, while LiFePO4/C material is relatively difficult. The impedance of the composite is lower than that of single LiFePO4/C after the two materials are combined. The results show that the addition of Li4Ti5O12 has a positive effect on the reduction of material resistance, improving the migration ability of Li+ and promoting the electrochemical performance.

Conclusions

Lithium-ion batteries have the advantages of high energy density, large charge and discharge current, and long cycle life, making them the most competitive and promising power source in this century. The improved method of this experiment effectively improves the electrochemical performance of LiFePO4.

The sample with the recombination ratio was 8:2 showed best electrochemical performance. The obtained composite material had fine particles and uniform size. In summary, compared with single LiFePO4, the electrochemical performance of LiFePO4/C-Li4Ti5O12 composites was superior, which promoted the development of LiFePO4 cathode materials and accelerates its practical development.