Introduction

Nowadays, the global energy crisis and environmental concerns are becoming more serious due to the continuous economic development and population growth. To mitigate these troubles, development and utilization of renewable energies (e.g., solar, wind and tidal energy) have become the focus of researches in many countries. Lithium-ion batteries (LIBs) (Zeng et al. 2019; Han et al. 2019) account for the main part in energy storage devices owing to their high energy density, reliability, long service life, and low environmental pollution (Xu et al. 2007). LIBs have been widely used in portable devices, transport systems like electric car (EVs) (Huang et al. 2018) and hybrid electric vehicles (HEVs) (Cusenza et al. 2019) as the most advanced energy storage equipments. The anode materials of LIBs are usually used as the main body for storing Li+, which directly determines the ability of lithium storage capability (Goriparti et al. 2014; Simon and Goswami 2011). Graphites are the traditional anode materials because of its high theoretical capacity (372 mAh g−1) (Zhang et al. 2019) and low cost. However, graphite’s shortages have restricted its further application in LIBs such as the formation of solid electrolyte interface (SEI) film (Huff et al. 2016), which seriously hindered ion transportation. A low working voltage (0.1 V) of the graphite brings capacity fading and fearful safety problems during the rapidLi+ lithiation/delithiation process. Therefore, it is very essential to develop new anode materials with higher stability and capacity in LIBs.

Except for traditional graphite anode materials, Li4Ti5O12 (LTO) has been widely studied by scientists in the past three decades (Liu et al. 2019a, b; Tojo et al. 2019), which is regarded to be the most representative titanates. The spinel LTO has a high embedding potential of 1.55 V (vs. Li/Li+) (Sandhya et al. 2014; Yi et al. 2014a, b) and can suppress the growth of lithium dendrites under high-rate charge/discharge. In addition, there is hardly volume change during the lithium insertion/extraction processes (Ahn and Xiao 2011). Consequently, LTO is a potential high-rate LIBs candidate. Nevertheless, the electronic conductivity of LTO is relatively low (Yuan et al. 2010) due to the 3D orbit of Ti element, which possibly leads to high polarization and capacities loss at higher cycle rates (Ouyang et al. 2007). In order to improve the electronic conductivity of LTO, three tactics are carried out as follows: (i) coating conductive materials on the surface of LTO (Xu et al. 2014; Wang et al. 2014); (ii) increasing the specific surface areas by changing the composites’ morphology to be nanofibers (Serife et al. 2015; Jung et al. 2015) or nanoparticles (Kamali and Fray 2015; Railey et al. 2017); (iii) doping some metal elements (Na+ (Yi et al. 2014a, b), Mg2+ (Lao et al. 2016), Co3+ (Wang et al. 2012), La3+ (Trong et al. 2015), Zr4+ (Hou et al. 2018), Ru4+ (Leng et al. 2019), Nb5+ (Tian et al. 2012), Sn2+ (He et al. 2017)) or non-metallic elements (F− (Zhao et al. 2013), Br− and W− (Zhang et al. 2015)) to stabilize the framework structures. The disadvantages of nonuniform coating thickness would decrease the capacities of the battery. Note the electrospinning technology has currently attracted extensive attention for one-dimensional nanofibers (Bhardwaj and Kundu 2010). Much more researchers prefer this technique to enlarge specific surface area with the purpose of improving the utility of the active materials. According to the previous studies, Li2CoTi3O8 nanofibers could be prepared by electrospinning technology to build the particular skeleton structures (Wang et al. 2012). In this paper, we adopt the means of electrospinning to fabricate a complex Li2CoTi3O8·CoTiO3·TiO2 (LCT), which is used for the anode materials in LIBs. TiO2 has played an important role in LIBs on account of its high structure stability, long cycle life, and prominent security (Cheng et al. 2020); the theoretical specific capacity of TiO2 is 335 mAh g−1. Meanwhile, CoTiO3 has been reported as one of the anode materials, which performs a function of enhancing cycling life (Liu et al. 2019a, b). So, we intend to add Co elements into Li4Ti5O12 to construct a novel kind of hierarchical nanohybrids LCT by electrospinning technology. According to last research studies, the theoretical specific capacity of Li2CoTi3O8 is calculated as 233.5 mAh g−1 (Wang et al. 2013). In addition, compared with the Li2CoTi3O8, the LCT as anode materials in LIBs display superior cycling stability and rate capability.

Experimental section

Preparation of materials

The LCT precursor composite nanofibers were fabricated through electrospinning, and then were treated by heating to get the final product. In a typical process, 0.4982 g cobalt tetrahydrate acetate (Macklin Chemical, 99.0%) was dissolved in 5 mL N,N-dimethylformamide (Macklin Chemical, 99.0%). In the meantime, 0.1320 g lithium acetate (Macklin Chemical, 99.9%) was added to the mixed solution of 5 mL absolute ethanol, 2 mL tetrabutyl titanate (Macklin Chemical, 99.0%), and 2 mL acetic acid (Macklin Chemical, 99.0%). The above solutions were mixed under vigorous magnetic stirring, and then 1.3802 g polyvinyl pyrrolidone (Aladdin Chemical, Mw = 1,300,000) was added into the hybrid solution with stirring constantly for 10 h to get the dark-red precursor solution for electrospinning.

The electrospinning experiments were carried out by loading the prepared mixture solution into a 10-mL plastic syringe with 17.0 kV and 15.0 cm of the vertical distance between the needle and the collector. The 0.6316 mL h−1 flow rate was employed at 35 °C. The nanofibers were collected into and dried at 80 °C for 4 h in a vacuum oven. The heated treatment included the preconditioning at 200 °C for 2 h with a heating rate of 2 °C min−1 and the calcining at 700 °C for 2 h with a heating rate of 3 °C min−1 under air atmosphere, which aimed to remove the organic polymers and obtain LCT composite nanofibers. As a contrast, the Li2CoTi3O8 nanofibers were prepared with the same stoichiometric ratio of cobalt tetrahydrate acetate, lithium acetate, and tetrabutyl titanate (Macklin Chemical, 99.0%) in the solvents. The schematic illustrations of the formation of LCT composite nanofibers and Li2CoTi3O8 nanofibers are shown in Fig. 1. All the chemical regents are of analytical grade in the experiment and used without further purification.

Fig. 1
figure 1

Schematic illustration of the preparation processes of the materials

Full size image

Characterization

The structures of the samples were detected by X-ray diffraction (XRD, Bruker D8 Focus Advance) using Cu-Ka radiation with scan angles ranging from 10° to 70°. The morphologies were characterized by scanning electron microscopy (SEM, Hitachi, SU-70) and high-resolution transmission electron microscopy (HRTEM, JEOL, JEM-2010) at an accelerating voltage of 200 kV. Energy-dispersive spectrometer (EDS) testing was performed in TEM mode. The distributions of valence states were conducted by X-ray photoelectron spectroscopy (XPS, Al Kα radiation). Thermogravimetric analysis (TGA, SIITG/DTA 6300) was analyzed under air atmosphere from 25 to 720 °C (5 °C min−1). The spinning process was accomplished by a FM1206 electrospinning machine (Beijing Fuyouma, China).

Electrochemical measurements

The CR2032-type coin cells were used to evaluate the electrochemical performances and the negative working electrodes consisted of active materials (LCT composite nanofibers or Li2CoTi3O8 nanofibers), carbon black and polyvinylidene fluoride (PVDF) binder in a mass ratio of 8:1:1 (wt%). After fully grinded, the electrode slurry was eventually coated on a Cu metal foils and dried at 100 °C for 10 h in the vacuum oven with the average loading mass for each electrode being 1.5 ± 0.2 mg cm−2. All the coin cells were assembled in an argon-filled glove box (H2O < 0.01 ppm, O2 < 0.01 ppm). Celgard 2400 and pure Li metal foils were used as separator and the counter electrodes, respectively. The electrolyte was a pre-made solution of 1.0 M LiPF6 solution in an ethylene carbonate (EC)/dimethyl carbonate (DMC) (1:1, v/v). Galvanostatic charge-discharge cycles were recorded on multichannel LAND-CT2001A battery test systems to measure coin cells at different current densities in the voltage range of 0.0~3.0 V. Electrochemical impedance spectroscopy (EIS, 0.01 ~ 100 kHz) and cyclic voltammetry (CV, 0.1 mV s−1, 0.0 ~ 3.0 V) were operated on a CHI660D electrochemical working station. All the electrochemical analyses were carried out at a relatively constant temperature of 25 °C.

Results and discussion

The schematic illustration of the electrospinning and subsequent calcination process of one-dimensional nanofibers are described in Fig. 1, in which the diameter of the LCT composite nanofibers is 250~300 nm (Fig. S1) before sintered and that of the nanofibers is 150 to 200 nm after sintered at 700 °C in an air atmosphere, the Li2CoTi3O8 nanofibers is 200~300 nm after sintered (Fig. S4). The high-resolution XPS spectra of the LCT composite nanofibers are exhibited in Fig. 2a–c to display the three different elements’ binding energies of Co 2p, Ti 2p, and O 1s. It can be seen from the Co 2p spectrum (Fig. 2a), the Co 2p peak splits into Co 2p3/2 (780.5 eV), Co 2p3/2 (781.8 eV), and Co 2p1/2 (796.8 eV) (Shi et al. 2020), respectively. In addition, there are also two satellite peaks at 786.6 and 802.5 eV (identified as “Sat.”), which have been proved to be the characteristic feature of high-spin divalent cobalt ions (Xu et al. 2015; Wang et al. 2020). Two peaks at 458.3 and 464.0 eV are correspond to Ti 2p3/2 and Ti 2p1/2 (Wang et al. 2013) in Fig. 2b. Meanwhile, the high-resolution spectrum displays the binding energy at 529.8 eV is fitted into O 1s (Fig. 2c). The EDS analysis of the LCT composite nanofibers in TEM mode is shown in Fig. 2d, which not only confirms the existence of Co, Ti, and O elements but also demonstrates the mass fraction and atomic fraction of the three elements in the attached table of the drawing. The atomic ratio of the elements Co:Ti:O is equal to 9.57:25.71:64.71. The Li element generally has no peak in EDS and other impurity elements like Cu and C are derived from the carbon film and copper mesh under TEM mode.

Fig. 2
figure 2

High-resolution XPS spectra of a Co2p, b Ti2p, and c O1s for the LCT composite nanofibers. d EDS spectra

Full size image

According to the XRD patterns of LCT composite nanofibers from Fig. 3a, several obvious diffraction peaks at 23.9°, 32.8°, 35.3°, 40.5°, 49.0°, 53.5°, 69.9°, and 63.5° correspond to the (012), (104), (110), (113), (024), (116), (214), and (300) lattice planes of CoTiO3 in JCPDS card no. 15-0866, respectively. The diffraction peaks at 14.9°, 18.3°, 23.7°, 26.0°, 30.1°, 35.5°, 43.2°, 57.1°, and 62.7° can be assigned to the (110), (111), (210), (211), (220), (311), (400), (511), and (440) of Li2CoTi3O8 (JCPDS card no. 49-0449), severally. And other diffraction peaks appearing at 27.4°, 36.1°, 41.2°, and 54.3° can be indexed to the (110), (101), (111), and (211) reflections of TiO2 (JCPDS no. 21-1276), respectively. For Li2CoTi3O8 nanofibers, all the diffraction peaks can be indexed to the Li2CoTi3O8 phase (JCPDS no. 49-0449) (Fig. 3b).

Fig. 3
figure 3

XRD patterns of LCT composite nanofibers (a) and Li2CoTi3O8 nanofibers (b). c TG-DTA curves of LCT composite nanofibers and d Li2CoTi3O8 nanofibers

Full size image

The TG curves of LCT composite nanofibers are depicted in Fig. 3c, in which four weight losses are able to be found during heating from 25 to 720 °C. The 22% of weight loss is related to the evaporation of absorbed water when the temperature rises to 100 °C. The second weight loss (38%) is accompanied by an exothermic peak near 305 °C in the DTA curve between 180 and 310 °C because of the decomposition of lithium acetate, cobalt acetate and tetrabutyl titanate. The weight loss of 10% in the range of 310~370 °C can be ascribed to the decomposition of inorganic salt with an exothermic weak peak that appears near 360 °C in the DTA curve. The weight has lost 17% in the temperature range of 370 ~ 440 °C, mainly caused by the thermal decomposition of the PVP. Upon further heating from 440 to 720 °C, there is no obvious weight loss indicating that the decomposition reaction is completed with the total weight loss of 83%. The TG profiles of Li2CoTi3O8 nanofibers are similar to LCT composite nanofibers (Fig. 3d), in which the weight loss of 20% in the range from 25 to 100 °C is related to the evaporation of absorbed water from the precursor fibers. The second weight loss (28%) between 180 and 305 °C is accompanied by an exothermic peak near 280 °C in the DTA curve because of the decomposition of lithium acetate, cobalt acetate, and tetrabutyl titanate. The third weight loss (18%) in the TG curve (305~370 °C) corresponds to the decomposition of inorganic salt appearing an exothermic weak peak near 314 °C in the DTA curve. The 17% of the last weight loss in the temperature changing from 370 to 410 °C suggests the thermal decomposition of the PVP. And then, no obvious weight loss from 410 to 720 °C shows the decomposition reaction is completed.

The morphology of LCT composite nanofibers is inspected by SEM (Fig. 4a, b), and the result shows that the LCT composite nanofibers have a uniform linear structure with the diameter of 150~200 nm after annealing. Transmission electron microscopy (TEM) is further implemented to verify the morphology and microstructure of the LCT composite fibers in Fig. 4c. Figure 4 d–f show that the crystallization interplanar spacings are closely related to the face (012) of CoTiO3, the face (111) of Li2CoTi3O8, and the face (111) of TiO2, respectively. As demonstrated in Fig. 4g, the distinct crystallization lattices indicate the CoTiO3 materials have well crystallization with the neighboring plane spacing of 0.3705 nm, which can be assigned to the spacing plane d012 of the CoTiO3 phase (JCPDS card no. 15-0866). Figure 4h displays that the prepared materials hold well crystallization with neighboring interplanar spacing of 0.4843 nm, which is assigned to the d111 spacing plane of the LiCoTi3O8 phase (JCPDS card no. 49-0449). The TiO2 materials perfectly crystallize with plane spacing of 0.2175 nm in Fig. 4i attributed to the d111 plane of the TiO2 phase (JCPDS card no. 21-1276). The above analysis results manifest that the composite nanofibers are a kind of complicated compound involving of CoTiO3, Li2CoTi3O8 and TiO2. The elemental mapping analyses of composite nanofibers are illustrated in Fig. 4j–m, which reveal that the elements of Co, Ti, and O are distributed homogeneously in the whole nanofibers.

Fig. 4
figure 4

a, b SEM images of the LCT composite nanofibers. c HRTEM image of composite nanofibers. d–f Interplanar crystallized spacing statistical tables. g–i HRTEM images. j–m Elemental mapping images of Co, Ti, and O

Full size image

The electrochemical cycling performances of Li2CoTi3O8·CoTiO3·TiO2 (LCT) composite nanofibers are described in Fig. 5a, in which the materials exhibit the first discharge capacity of 351.41 mAh g−1 at the current density of 100 mA g−1. After 120 cycles of charge and discharge, the reversible lithium storage capacity of the LCT composite nanofibers has decreased to 201.68 mAh g−1. The coulombic efficiency is close to 100% during the first 120 cycles of charge and discharge. As a comparison, the initial capacity of Li2CoTi3O8 nanofibers achieves 408.88 mAh g−1 at the current density of 100 mA g−1, while its reversible lithium storage capacity rapidly decreases to 90.23 mAh g−1 after 100 cycles (Fig. S2a). It clearly indicates that the LCT composite nanofibers hold higher reversible lithium storage capacity than the Li2CoTi3O8 nanofibers. Therefore, it demonstrates that the unique composite structure is significant to enhance the rate capability of the materials.

Fig. 5
figure 5

a The cycling properties of LCT composite nanofibers at current densities of 100 mA g−1. b Cyclic voltammograms curves at a scan rate of 0.1 mV s−1 from 0.01 to 3.00 V. c Charge/discharge curves of at current densities of 100 mA g−1. d Rate capabilities of. LCT composite nanofibers. e Charge/discharge curves at current densities of 1000 mA g−1. f EIS patterns for the LCT composite nanofibers and Li2CoTi3O8 nanofibers

Full size image

As shown in Fig. 5b, cyclic voltammetry (CV) tests are carried out to evaluate the electrochemical activity of the LCT composite nanofibers. The interactions between with Li+, Li2CoTi3O8, CoTiO3, and TiO2 could be depicted by the following reactions (Wang et al. 2012; Jeong et al. 2014; Liu et al. 2017):

$$ {\left[{\mathrm{Li}\mathrm{Co}}_{0.9}\right]}^{\mathrm{tet}}{\left[\left({\mathrm{Li}\mathrm{Co}}_{0.1}\right){\mathrm{Ti}}_3\right]}^{\mathrm{oct}}{\mathrm{O}}_8+{3\mathrm{e}}^{-}+{3\mathrm{Li}}^{+}\leftrightarrow {\left[{\mathrm{Li}}_4\right]}^{\mathrm{oct}}{\left[{\mathrm{Co}}_{0.9}\right]}^{\mathrm{tet}}{\left[\left({\mathrm{Li}\mathrm{Co}}_{0.1}\right){\mathrm{Ti}}_3\right]}^{\mathrm{oct}}{\mathrm{O}}_8 $$
(1)
$$ {\mathrm{TiO}}_2+{\mathrm{x}\mathrm{Li}}^{+}+{\mathrm{x}\mathrm{e}}^{-}\leftrightarrow {\mathrm{Li}}_{\mathrm{x}}{\mathrm{TiO}}_2\ \left(0\le \mathrm{x}\le 1\right) $$
(2)
$$ {\mathrm{CoTiO}}_3+{\mathrm{x}\mathrm{Li}}^{+}+{\mathrm{x}\mathrm{e}}^{-}\leftrightarrow \mathrm{Co}+{\mathrm{Li}}_{\mathrm{x}}{\mathrm{TiO}}_2+{\mathrm{Li}}_2\mathrm{O} $$
(3)

During the first three cycles, the oxidation peak at 1.76 V indicates the process of LIBs’ delithiation and two obvious reduction peaks at approximately 0.18 V and 0.17 V correspond to the lithium-ion intercalation process. For Li2CoTi3O8 nanofibers, the oxidation peaks at 1.75 V and 0.16 V and reduction peak at about 0.3 V are shown in Fig. S3b.

It can be seen from Fig. 5c that the LCT composite nanofibers display the representative voltage profiles of charge/discharge curves at the current density of 100 mA g−1 with the potentials ranging from 0.0 to 3.0 V. The LCT composite nanofibers deliver an initial discharge capacity of 351.41 mAh g−1 and an initial charge capacity of 201.68 mAh g−1, exhibiting superior electrochemical performances than other Li2CoTi3O8-based composite materials (Table S1). After the first cycle, the irreversible capacity loss of 149.73 mAh g−1 mainly originates from the formation of the solid electrolyte interphase (SEI) film (Huff et al. 2016). At the same circumstances, the charge/discharge voltage profiles of Li2CoTi3O8 nanofibers perform the initial discharge capacity of 408.88 mAh g−1 and initial charge capacity of 181.64 mAh g−1 with the irreversible capacity loss of 227.24 mAh g−1 (Fig. S3a).

Figure 5d shows the rate performances of the LCT composite nanofibers at different current densities from 100 to 800 mA g−1. The initial reversible capacity of the composite nanofibers is 196.45 mAh g−1 at the current density of 100 mA g−1. Subsequently, the charge capacities reduce slowly to 171.36, 148.35, 134.14, 121.67, 110.63, 101.57, and 97.14 mAh g−1 when the current densities gradually increase to 200, 300, 400, 500, 600, 700, and 800 mA g−1, respectively. Upon the current density turns back to 100 mA g−1, the charge capacity of the composite nanofibers can still recover to 185.59 mAh g−1. By contrast, the rate performances of Li2CoTi3O8 nanofibers are revealed in Fig. S2b, where Li2CoTi3O8 nanofibers present low charge capacities of 130.15, 120.27, 111.35, 102.34, 97.83, 95.10, 90.12, 85.38, 80.16, and 141.13 mAh g−1 at the current densities of 50, 100, 200, 300, 400, 500, 600, 700, 800, and 50 mA g−1, respectively. As shown in Fig. 5e, when the high current density of 1000 mA g−1 is exerted, the capacity of the composites gradually increases from 61.72 to 110.35 mAh g−1 before 300 cycles, and then decreases to 62.39 mAh g−1 in the next 500 cycles, which can be regarded as the reason of the activation process of the materials. All the above special rate performances may result from the following aspects: (i) the one-dimensional (1D) nanostructures provide sufficiently large interface areas between electrodes and electrolyte for the Li+ rapid transfer during lithiation/delithiation; (ii) the advantageous structures of nanocomposites relieve volume changes and pulverization of the materials during cycling; (iii) the ternary composite materials Li2CoTi3O8·CoTiO3·TiO2 can play a synergistic role in improving electrochemical properties effectively.

To further improve electrochemical performances, electrochemical impedance spectroscopy (EIS) analyses of the nanofibers are performed at room temperature, of which the EIS curves are shown in Fig. 5f. The intercepts of the Nyquist plots at the high-frequency regions can be attributed to the bulk resistance of the liquid electrolyte (Re), whereas the semicircle in the medium frequency regions is the charge-transfer resistance (Rct) of the electrode/electrolyte interfaces. Meanwhile, the line in the low-frequency region corresponds to the diffusion limitation within the electrodes. CPE is defined as a constant phase element, and W0 represents Warburg impedance. The equivalent circuit diagram can perfectly match the EIS test results (Fig. 5f). The fitting results are shown in Table 1, where it has been observed that the Rct of LCT composite nanofibers (26.91 Ω) is much smaller than that of Li2CoTi3O nanofibers (41.35 Ω). Moreover, the Rs of LCT composite nanofibers (7.95 Ω) is also lower than that of Li2CoTi3O nanofibers (9.18 Ω), indicating that the electrical conductivities of LCT composite nanofibers are superior to that of Li2CoTi3O nanofibers.

Table 1 The fitting results of LCT composite nanofibers and Li2CoTi3O8 nanofibers
Full size table

Conclusions

In summary, LCT composite nanofibers were synthesized feasibly by the electrospinning method and characterized by XRD, TG-DTA, SEM, TEM, XPS, EDS, CV, and so on. After 120 cycles, the tests show that LCT composite nanofibers as LIBs anode materials can retain a reversible capacity of 201.68 mAh g−1 at the current density of 100 mA g−1. The electrochemical performances exhibit large cycle capacity and high rate capability, in which the combined composite materials can achieve superior performances. This work provides splendid ideas for the fabrication of different composite materials, and the synthesized LCT composite nanofibers can be considered as promising candidates for optimal anode materials for LIBs.