1 Introduction

Emergence of lithium ion batteries (LIBs) as a secondary energy storage technology can be attributed to their high energy density, better compatibility and lightness [1]. In order to equip LIBs for high-end applications like electric vehicles, plug in hybrid vehicles and other hybrid devices, its prime component, the electrode materials should have improved properties so as to deliver high performance. Intercalation based anode materials received much attention as LIB electrodes due to their low cost, easy accessibility, better reversibility and low working potential. Even though the first commercial anode material for LIB—graphite, is still in market, better alternatives are still under exploration [2]. Principal shortcoming of graphite is its lower intercalation potential which could lead to dendrite formation further leading to unfavourable reactions and cell failure. Additionally, lithium plating restricts its high rate performances owing to its polarization effects. Titanates are titanium based inorganic compounds specifically, lithium titanate- (Li4Ti5O12-LTO), sodium titanate (Na2TiO7) and the likes are quite a while explored as LIB anodes. Among them, LTO has already been commercialized in LIBs due to its attractive properties like environmentally benign, negligible volume expansion and good reversibility [3]. Due to high voltage discharge cut-off of LTO (1.0 V vs. Li/Li+), electrolyte decomposition is limited and no significant solid electrolyte interface (SEI) layer is expected to be formed leading to high Coulombic efficiency. Nevertheless its inherently low ionic and electronic conductivity is still a challenge. Utilization of numerous approaches like doping, compositing with carbon based materials, morphology control and compositing with other metal oxide have been widely explored successfully [4,5,6,7,8]. Compositing TiO2 with LTO is considered to be an effective strategy to improve the conductivity and kinetic issues of LTO [9,10,11,12,13]. Recently, we investigated the LTO-TiO2 complete range of composite from 0 wt.% of TiO2 (i.e. LTO = 100 wt.%) to 100 wt.% of TiO2 (i.e. LTO = 0 wt.%) and confirmed that TiO2 15 wt.% (LTO 85 wt.%) performed the best compared to all the ratios systematically investigated [14]. Yet, long cycling stability and high specific capacity at high rates along with its practical use in the form of full-cells is important to be reported. In this work, the best performing LTO-TiO2 composite ratio of 85–15 wt.% was prepared and its electrochemical performance for two different end applications namely, (i) Li-ion battery and (ii) Li-ion capacitor is reported. The composite nanoparticles upon testing as half-cells for lithium ion battery exhibited ultra-long cycling stability and high capacity, making it an ideal negative electrode material for energy storage applications. The excellent high rate electrochemical performance of LTO-TiO2 electrode fabricated as negative electrode in lithium ion battery and lithium ion capacitor (LIC) further confirmed the potential of the composite.

2 Materials and methods

Titanium isopropoxide (Sigma-Aldrich, USA), lithium hydroxide monohydrate (Sigma-Aldrich, USA), ammonium hydroxide (Nice chemicals), ethylene glycol (Merck, India) were used as raw material for synthesis of LTO nanoparticles. Solvothermal process was adopted to obtain the composite nanoparticles using lithium deficient stoichiometric precursors that were mixed (to obtain LTO-TiO2 composite nanoparticles), followed by the drop wise addition of 4 ml ammonia solution [14]. As obtained solution was transferred to a Teflon autoclave and maintained at 180 °C for 36 h. Finally solution was washed and dried before annealing at 500 °C for 6 h.

Crystal structure and purity of sample was confirmed by X-ray diffraction, XRD (Rigaku ultimate-IV, Japan) while morphological analysis was recorded using Transmission electron microscopy, TEM (FEI TECNAI G2, USA) operating at 200 kV voltage. Electrodes for were fabricated by slurry casting technique by blending active material, conductive carbon (either carbon black or CNT) and binder (PVdF) in the ratio 75:15:10. Coin cells or Swagelok cells were fabricated inside glove box using 1 M LiPF6 in EC:DMC (1:1) as the electrolyte for the present work. Electrodes were tested in battery cycler either in Arbin instruments (USA) or in BioLogic Instruments (USA). Half-cells were fabricated with lithium metal as counter electrode and tested in the voltage window of 1.0 V to 2.5 V (with the electrode loading of about 1–1.5 mg/cm2). Lithium ion battery was tested by coupling LTO-TiO2 with LiMn2O4 cathode synthesized in our lab while lithium ion capacitor was fabricated by coupling LTO-TiO2 with activated carbon electrodes obtained from a commercial supercapacitor. The activated carbon electrode half-cells were tested with a loading of 8–10 mg/cm2. For the capacitor application, LTO-TiO2 electrode was pre-conditioned at 1C rate (175 mA/g) for one cycle against lithium metal. Subsequently the LTO-TiO2 half-cell was disassembled and coupled with activated carbon for fabricating a lithium ion capacitor device.

3 Results and discussion

The LTO-TiO2 composite material’s structure and morphology was characterized by XRD and TEM analysis. Due to the lithium-deficient stoichiometry of precursors, XRD pattern (Fig. 1a) shows the co-existence of dual-phase cubic spinel LTO and tetragonal TiO2. From the relative intensity ratios, the weight percentage of LTO and TiO2 was calculated to be 86% and 14% respectively [14, 15]. To quantify the LTO-TiO2 composite ratio XRD is ideal as it is difficult to quantify using other techniques as Ti and O are common in both the phases while Li being only element different does not provide significant difference in most of the techniques. Figure 1b and c shows low magnification and high resolution TEM images of the synthesized nanoparticles. Low magnification image shows almost uniform sized nanoparticles (average size of about 30–35 nm) while HRTEM confirms the crystallinity of the particle. LTO particles are oriented along (111) with lattice spacing of 0.49 nm while TiO2 particles along (004) with 0.24 nm spacing. TiO2 orientation along (004) is proved to be beneficial in improving the lithiation kinetics than the usual (101) plane [13, 15]. The Brunauer–Emmett–Teller, BET surface area of these composite nanoparticles were measured to be 64 m2/g and the isotherm is depicted in Fig. 1d. Such a high surface area nanoparticles leads to good particle–particle electrical contact and electrode–electrolyte contact in the electrode.

Fig. 1
figure 1

a X-Ray diffraction pattern b Low magnification TEM image c HRTEM and d BET adsorption isotherm of composite LTO-TiO2 nanoparticles

Full size image

Figure 2a shows the first three cycle’s cyclic voltammogram of composite LTO-TiO2 nanoparticles in the potential window ranging from 1.0 V to 2.5 V with 0.1 mV/s as scan rate. As can be noted, two significant redox peaks at 1.45 and 1.62 V corresponds to the Li4Ti5O12/Li7Ti5O12 transition during the redox reaction while the minor peaks at 1.74 V and 1.97 V correlated to TiO2/LixTiO2 transition upon lithiation and delithiation. The occurrence of exactly overlapped current–voltage curves indicate the excellent reversibility of the electrode. Charge–discharge cycling was conducted at constant current mode so as to evaluate the electrochemical performance of LTO based electrode. Figure 2b shows the 1st cycle charge–discharge profile of the electrode at 1C rate (1C = 175 mAh/g). In accordance with the CV profile we can observe plateaus emerging from the lithiation reaction of LTO and TiO2 as distinct plateaus. The electrode delivered first cycle discharge capacity of 203 mAh/g at 1C rate. This value is higher than the theoretical capacity of LTO because TiO2 is also electrochemically active in the same potential window (1 V to 2.5 V) contributing the additional capacity. The enhanced capacity can be attributed to the capability of nanosized high surface area material to provide better electrolyte/active material contact and transport through LTO-TiO2 interface promoting faster Li-ion transport. The same upon cycling (Fig. 2c), displayed high specific capacity (172 mAh/g) and retention (89%) even at the end of 100th cycle.

Fig. 2
figure 2

Electrochemical characterization results of the composite nanoparticles a cyclic voltammetry profile b 1C charge–discharge profile and c 1C cycling for 100 cycles

Full size image

It may be noted that at low rates the capacity may be dominated with intercalation yet, at high rates the capacity may have contributions from both intercalation and capacitive storage (as the surface area of the composite is high). To evaluate the long cycling performance of composite electrode, we have conducted galvanostatic charge–discharge test at high (2000 mA/g) C-rates. Figure 3 shows cycling performance and Coulombic efficiency data. The electrode exhibited high specific capacity of 168 mAh/g in the first cycle and cycling stability over 5000 cycles. Initial Coulombic efficiency was observed as 92.4% subsequently, the Coulombic efficiency improved and stabilized above 99% and the electrode also retained 77% of charge capacity at the end of 5000 cycles. This ultra-long cycling, high rate performance of LTO-TiO2 electrode was observed to be better than many literatures with pristine, composited, coated and doped LTO structures [16,17,18,19,20,21,22]. Thus the unique combination of LTO-TiO2 composite electrode delivered high specific capacity, excellent cycling stability and Coulombic efficiency, making it an appealing electrode for lithium ion storage applications. Good electrochemical performance of LTO-TiO2 composite is due to the LTO-TiO2 interface assisted Li-ion intercalation, CNT as conductive additive and smaller particle size leading to shorter diffusion length. All these aspects help in improving both electronic conductivity and Li-ion diffusion kinetics enabling high rate capability.

Fig. 3
figure 3

Ultra-long cycling performance of the composite nanoparticles at a current rate of 2.0 A/g

Full size image

Inspired by the cycling performance of LTO-TiO2 half-cells, full cell Li-ion batteries were fabricated and tested. For full cell device, LTO-TiO2 anode was coupled with electrode with LiMn2O4 cathode, synthesized in our lab [23]. Electrodes loading were optimized with slightly cathode dominated cell (capacity p/n = 1.15) and tested for cycling at 1770 mA/g (~ 10.1 C) based on the anodes active weight. Figure 4a shows the 1st and 1000th cycle charge–discharge profiles of the 2.35 V Li-ion battery. The cell was able to deliver an initial discharge capacity of 126 mAh/g and an average Coulombic efficiency of 88%. The energy density delivered for the first cycle was calculated based on the delivered capacity and average voltage of the full-cell which was 126 × 2.35 = 296 Wh/kg. Based on the energy density and time of discharge (4.35 min = 0.0725 h), the first cycle power density of the full-cell was calculated as 296/0.0725 = 4082 W/kg. The energy density and power density calculated for the full-cell tested at a specfic current of 1770 mA/g based on anodes active weight. Additionally, the cell delivered cycling performance for 1000 cycles with capacity retention of around 75% (Fig. 4b).

Fig. 4
figure 4

a Charge–discharge profiles of LTO-TiO2/LMO cell and b long cycling performance for 1000 cycles

Full size image

The potential of LTO/TiO2 electrode was studied by fabricating a LIC with LTO/TiO2 composite as negative electrode and activated carbon electrode as positive electrode, results of which are discussed below. The activated carbon electrode half-cell performance is presented in Fig. 5. To fabricate the LIC, activated carbon (AC) electrode retrieved from a commercial double-layer symmetric supercapacitor device (by disassembling the electrode subsequenlty washed and dried). The AC that exhibit electrochemical half-cell performances displayed in Fig. 5 was used as positive electrode against LTO for Li-ion capacitor (LIC). The AC electrode (against Li-metal with organic electrolyte) was able to deliver a specific capacitance of 150 F/g at a current rate of 100 mA/g (Fig. 5a) and about 89 F/g at a current rate of 1000 mA/g (Fig. 5b) at the end of 150 and 1000 cycles respectively. The LIC was cycled at a faster rate of 10,000 mA/g (~ 57 C) in the potential window ranging from 1.0 V to 3.0 V. As mentioned in the experimental section, the pre-conditioned negative electrode is coupled with the activated carbon electrode. The capacitor was optimized based on the performance of half-cells. Various weight ratios of negative to positive electrode were tested and it was observed that 1:10 weight ratio performed better and thus utilized for final device testing. Voltage profiles of the cell at different cycle number is displayed in Fig. 6a and b while, Fig. 6c shows the long cycling performance of the cell for 10,000 cycles at 10 A/g specific current. The cell was able to deliver initial discharge capacitance of 151.8 F/g and irrespective of the pre-conditioning the negative electrode, the first cycle Coulombic efficiency was only around 45% and requires further optimization. In subsequent cycles, the capacitance and Coulombic efficiency increases and reaches maximum capacitance over 250 F/g and 100% respectively. The initial increase in the capacitance and subsequent decrease is similar to the activated carbon results presented in Fig. 5b. The capacitance increase indicates that at high rates several sites get activated during initial stages of cycling. The cell was able to cycle for around 10,000 cycles and with the final discharge capacitance value of 196 F/g. The LIC performance was limited by the activated carbon electrode and thus can be improve further by utilizing high surface area activated carbon electrode. The excellent high rate cycling performance of lithium ion battery and lithium ion capacitor demonstrate the potential of composite LTO-TiO2 for practical applications.

Fig. 5
figure 5

Galvanostatic charge–discharge cycling of activated carbon at a 100 mA/g and b 1000 m

Full size image
Fig. 6
figure 6

Galvanostatic charge–discharge profiles a, b of the lithium ion capacitor at current rate of 10,000 mA/g and (c) long cycling performance of the same for 10,000 cycles

Full size image

4 Conclusions

In conclusion, we reported the electrochemical performance of composite LTO-TiO2 nanoparticles prepared through a solvothermal synthesis. The material was observed to be highly crystalline with individual particles of below 30 nm in size and BET surface area of 64 m2/g. Electrochemical investigation of as synthesized LTO-TiO2 composite direct to the fact that these nanoparticles have excellent specific capacity, cycling stability and retention. Both high and low rate performances were demonstrated to confirm the potential of these nanoparticle electrodes. This study reveals an effective strategy to improve the performance of LTO by developing an LTO based material which could deliver enhanced specific capacity, rate performance and excellent cyclic stability even for 5000 cycles. Additionally, lithium ion battery (1000 cycles at 1000 mA/g) and lithium ion capacitor (10,000 cycles at 10,000 mA/g) cycling performance confirms the potential of the material for high power practical applications.