Graphene-wrapped Li4Ti5O12 hollow spheres consisting of nanosheets as novel anode material for lithium-ion batteries

Graphene-wrapped Li4Ti5O12 hollow spheres consisting of nanosheets as novel anode material for lithium-ion batteries

Electrochimica Acta 254 (2017) 287–298 Contents lists available at ScienceDirect Electrochimica Acta journal homepage: www.elsevier.com/locate/elect...

5MB Sizes 262 Downloads 75 Views

Electrochimica Acta 254 (2017) 287–298

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Research Paper

Graphene-wrapped Li4Ti5O12 hollow spheres consisting of nanosheets as novel anode material for lithium-ion batteries Zhiya Lina,c , Yanmin Yanga,b , Jiamen Jina , Luya Weia , Wei Chena , Yingbin Lina,b,c,* , Zhigao Huanga,b,c a College of Physics and Energy, Fujian Normal University, Fujian Provincial Key Laboratory of Quantum Manipulation and New Energy Materials, Fuzhou, 350117, China b Fujian Provincial Engineering Technology Research Center of Solar Energy Conversion and Energy Storage, Fuzhou, 350117, China c Fujian Provincial Collaborative Innovation Center for Optoelectronic Semiconductors and Efficient Devices, Xiamen, 361005, China

A R T I C L E I N F O

Article history: Received 7 July 2017 Received in revised form 9 September 2017 Accepted 20 September 2017 Available online 22 September 2017 Keywords: lithium-ion batteries lithium titanium oxide graphene diffusion coefficient Work function

A B S T R A C T

Flower-like Li4Ti5O12 hollow microspheres consisting of nanosheets are prepared via a hydrothermal process, and subsequently wrapped by graphene through electrostatic interactions. In comparison with pristine Li4Ti5O12, [email protected] exhibited higher capacities and improved rate capability in the 0.01–3.0 V or 1.0–3.0 V potential range. [email protected] composite shows specific capacity of 272.7 mAh g1 at 750 mA g1 after 200 cycles in the potential range from 0.01 to 3.0 V, while the pristine Li4Ti5O12 only delivered a discharge capacity of 235.6 mAh g1. The improved electrochemical performances of [email protected] should be attributed to lower charge-transfer resistances, larger lithium-ion diffusion coefficient and lower activation energy. The electrons transfer at Li4Ti5O12/ graphene heterojunction interface, originating from difference in the work function of two composites, reduces the localized work function of the composites, decreases energy required for electrons to escape and consequently results in the improved electrochemical performances. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction In the past two decades, lithium-ion batteries with high energy density have attracted increasing attention due to their potential application in large-scale energy storage, electric vehicles and hybrid electric vehicles [1–3]. Among numerous anode materials, spinel lithium titanate oxide (Li4Ti5O12) has been regarded as a promising alternative to the carbon-based anode materials because of its relatively high li-insertion potential (1.55 V vs. Li/ Li+), excellent thermal stability and high lithium-ion insertion/ extraction reversibility [4–6]. Unfortunately, Li4Ti5O12 has the intrinsically slow lithium-ion diffusion coefficient and poor electronic conductivity, which results in severe electrode polarization, poor kinetics of conversion reaction and seriously limits its high-rate performance in practical applications. To address these issues, lots of effective strategies have been developed including coating with conducting layer, designing hierarchically structure

* Corresponding author at: College of Physics and Energy, Fujian Normal University, Fujian Provincial Key Laboratory of Quantum Manipulation and New Energy Materials, Fuzhou, 350117, China. Tel.:+ +86 591 2286 8132 fax: +86 591 2286 8132. E-mail address: [email protected] (Y. Lin). https://doi.org/10.1016/j.electacta.2017.09.123 0013-4686/© 2017 Elsevier Ltd. All rights reserved.

and doping with isovalent ions [7–11]. Comparing to the bulk Li4Ti5O12 synthesized at high calcination temperature, micro-nano hierarchical structure assembled by nanounits should be more desirable for faster lithium insertion/extraction kinetics because of the shorter Li + diffusion length. However, numerous active sites and relatively lower degree of crystallinity of the nano-structured Li4Ti5O12 easily results in the poor stability of crystal structure, the large resistivity of the electrode and consequently deterioration of electrochemical performances [12]. A conductive carbon coating or addition has been proven to be the most effective way in improving the rate capacity and cycling stability of Li4Ti5O12 battery, which could significantly enhance the surface electrical conductivity and the electrical contact with electrolyte solution [13]. Mixing lithium and titanium precursor with carbon precursor and subsequently thermal-annealing in an inert to form a carbon-coating layer, is well known as a traditional way to introduce carbon-coating layer on Li4Ti5O12 surface [14,15]. It should be pointed out that higher electronic conductivity of carbon-coating layer always requires higher carbonization temperature. High carbonization temperature would cause the loss of oxygen in Li4Ti5O12 and agglomeration of particles, which has a significant adverse effect on the electrochemical performances. Hybridization of Li4Ti5O12 with graphene has been paid increasing attention because graphene has

288

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

excellent electronic conductivity, large specific surface area and high mechanical strength. Hybridization with graphene can not only enhance electrical conductivity, but also improve the cyclic stability and rate performance of anode materials. In our work, Li4Ti5O12 hollow microspheres consisting of nanosheets are wrapped by graphene through electrostatic interactions and subsequently thermal-annealed in the moderate temperature, which could not only maintain the nanosheetstructure characteristics of Li4Ti5O12, but also form a highly conductive network for electron transportation in the composites. In comparison, [email protected] electrodes exhibit higher discharge capacity and superior rate capability. The improved kinetics of Li+ insertion/extraction by graphene-wrapping are systematically investigated. 2. Experimental 2.1. Synthesis of Li4Ti5O12 microspheres Flower-like Li4Ti5O12 hollow microspheres consisting of nanosheets are prepared by a hydrothermal method using lithium hydroxide monohydrate (LiOHH2O) and tetrabutyl titanate (Ti (OC4H9)4) as precursors. In a typical process, 16 mmol LiOHH2O is dissolved in 30 ml 30 wt.% H2O2 and 40 mL de-ionized water under stirring for 30 minutes. Then, 1.2 ml Ti(OC4H9)4 is added dropwise into the above aqueous solution during continuous stirring. The resulted suspension is transferred into a Teflon-lined autoclave of

100 mL capacity, sealed and maintained at 150  C for 6 hours. The obtained precipitates are subsequently collected via centrifugation, and further washed with absolute ethylalcohol several times followed by drying at 80  C in vacuum. Finally, the as-prepared lithium titanium oxide precursors are sintered at 500  C for 5 hours in air to get Li4Ti5O12 powders. 2.2. Synthesis of [email protected] [email protected] composites are synthesized by electrostatic interactions between negatively charged graphene oxide sheets and positively charged Li4Ti5O12 particles in aqueous solution. Typically, 0.1 g of Li4Ti5O12 microspheres are dispersed in 200 mL de-ionized water under ultrasonication for 30 minutes, followed by adding 1 mL of 3-aminopropyltriethoxysilane (APTES) under strong stirring. After the mixture is stirred for 12 hours, the resulted precipitates are collected by centrifugation and washed with ethanol and de-ionized water several times to obtain aminosilane modified Li4Ti5O12 nanoparticles. The APTES induces positive charges on the Li4Ti5O12 surface. The graphene-wrapped Li4Ti5O12 composites are prepared by electrostatic interactions between positively-charged amino-silane modified Li4Ti5O12 and negatively-charged graphene oxide. For the selfassembly, 2L 2 M graphene-oxide suspension is added dropwise into the aminosilane modified Li4Ti5O12 dispersion under vigorous stirring for 3 hours, resulting in the Li4Ti5O12 particles are strongly adsorbed on to the graphene-oxide surface via electrostatic interactions.

Fig. 1. (a) XRD patterns of Li4Ti5O12 and [email protected] powders; (b) TG curve of the [email protected] at 5  C min1 in air; (c) Nitrogen sorption isotherms and (d) pore diameter distribution of Li4Ti5O12 and [email protected] graphene powders.

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Then, the precipitate is collected by centrifugation, washed with de-ionized water several times and sintered at 500  C in an argon atmosphere for 3 hours to obtain the final product. 2.3. Materials characterization The crystalline phase of the as-prepared composites was investigated by powder X-ray diffractometry (XRD, Rigaku MinFlex II) with Cu-Ka radiation (l = 0.15406 nm). The morphologies of the Li4Ti5O12 and [email protected] are analysed by field-emission

289

scanning electron microscopy (FESEM; HITACHI, SU-8010) and transmission electron microscopy (TEM, Tecnai G2 F20 S-TWIN). The graphene content in the composite is determined by thermogravimetric analysis (TGA, Netzsch STA449F3) at a heating rate of 5  C min1 under an air atmosphere. Nitrogen sorption isotherms are measured at 77 K using a Micromeritics Tristar 3020 analyzer. Specific surface areas of the as-synthesized powders are calculated according to the Brunauer–Emmett–Teller (BET) method. The pore size distribution is determinated based on the theory of Barrett Joyner and Halenda (BJH). The surface potentials of

Fig. 2. (a,b) SEM images of the as-prepared Li4Ti5O12 powders; (c) SEM images of [email protected] powders; (d,e) HR-TEM images of [email protected] powders.

290

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Li4Ti5O12 and [email protected] electrodes are measured by Kelvin probe atomic force microscopy (KPAFM) (Bruker dimension ICON, Germany). 2.4. Cell fabrication and characterization The electrochemical performances are evaluated using CR2015 coin cells with the (Li4Ti5O12 or [email protected]) anode electrode coated on copper foil and lithium foil as a counter electrode. The anode slurry is prepared by mixing 70 wt.% active material (Li4Ti5O12 or [email protected] graphene) with 10 wt.% polyvinylidene fluoride (PVDF) and 20 wt.% super-P in a solvent (Nmethy1-2-pyrrolidone). The slurry is then cast onto a copper foil and subsequently dried at 110  C in vacuum for 12 hours. The resulted laminate is punched into a 12.5 mm diameter disc and pressed under a pressure of 2 MPa. The loading density of the active materials is 1.3 mg Cm2. Electrochemical cells are assembled in an Ar-filled glove box (O2 <0.5 ppm, H2O <0.5 ppm) using a microporous polypropylene membrane (Celgard 2400) served as a separator and 1 M LiPF6 in dimethyl carbonate(DMC)-ethylene carbonate (EC) (1:1 in volume) as the electrolyte. The galvanostatic charge/discharge measurements are carried out on a LAND test system (LAND CT2001A). The cyclic voltammograms are collected at a scan rate of 0.1 mV s1 on an electrochemical workstation (Arbin Instruments BT2000, USA). Electrochemical impedance spectra of the cells are recorded in the frequency range of 100 KHz– 10 mHz with an AC amplitude of 5 mV on a Zahner Zennium IM6 electrochemical workstation.

Fig. 3. The rate capability of Li4Ti5O12 and [email protected] graphene electrodes.

between the dark edge and the gray center of microsphere, revealing the hollow nature of the as-prepared Li4Ti5O12 microsphere. HRTEM image in Fig. 2(e) further confirms that the uniformity coverage of the graphene nanosheets on Li4Ti5O12 surface and the strong adhesion between graphene sheets and

3. Results and discussion Fig. 1(a) shows XRD patterns of the as-synthesized Li4Ti5O12 and [email protected] composites. All the diffraction peaks of all samples could be assigned typically to a cubic structure of spinel Li4Ti5O12 (JCPDS No. 49-0207) and no other phase is detected, indicating graphene-wrapping does not change the crystal structure of spinel Li4Ti5O12 during the calcination process. The results of thermal analysis (TG/DTA) on the [email protected] reveals that the graphene content in composite is about 5.0 wt%, shown in Fig. 1(b). Fig. 1(c) shows nitrogen adsorption-desorption isotherms of Li4Ti5O12 and [email protected], suggesting a typical hysteresis mesoporous system [16–18]. Based on Brunauer–Emmett–Teller (BET) analysis, the specific surface areas of Li4Ti5O12 and [email protected] are calculated to be 78.86 and 108.38 m2g1, respectively. Large specific surface area undoubtedly provides a large contact area with the electrolyte and leads to high li-ion flux across the electrolyte/electrode interface [19–21]. The Barrett–Joyner–Halenda (BJH) pore-size distribution shown in Fig. 1(d), reveals Li4Ti5O12 and [email protected] have an average pore size of 21.86 nm and 15.85 nm respectively. An increase in surface area and reduction in average pore size by graphenewrapping understandably result from the high specific surface area and porosity of graphene. Fig. 2 demonstrates the typical SEM images of the hierarchical Li4Ti5O12 and [email protected] powders. As shown in Fig. 2(a), a typical product is consist of numerous flower-like Li4Ti5O12 clusters. Fig. 2(b) is a magnified SEM image of flower-like Li4Ti5O12 clusters, showing that flower-like hierarchical nanostructure has hollow structure and is composed of numerous nanosheets with approximately 100 nm wide and 250 nm long. Such hierarchical structure should be highly desirable for rapid charge transfer and li-ion diffusion. Fig. 2(c) shows that Li4Ti5O12 microspheres are wrapped by gauze-like graphene nanosheets and the adjacent Li4Ti5O12 microspheres are connected by wrinkled graphene sheets, which provides a highly conductive network among the insulating Li4Ti5O12 particles. Fig. 2(d) shows the obvious contrast

Fig. 4. (a) CV curves of Li4Ti5O12 and [email protected] electrodes at a scanning rate of 0.1 mV s1; (b) AC impedance spectra of Li4Ti5O12 and [email protected] electrodes at the stable voltage of 1.55 V.

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Li4Ti5O12 particles. It is believed that the strong adhesion between Li4Ti5O12 and graphene promotes electrons migration and thus enhances electronic conductivity of the composite, as well as preventing aggregation of Li4Ti5O12 particles during the lithiumion insertion/deinsertion proces [22]. Li4Ti5O12 particles in turn prevent the aggregation of graphene sheets, which maintains high specific surface area of graphene and provides additional intercalation sites. In addition, the distinct parallel fringes with a basal spacing value of approximately 0.48 nm are observed, correspongding to the (111) crystalline plane of Li4Ti5O12 and indicating the high crystallinity of Li4Ti5O12 phase. Fig. 3 shows the discharge capacity as a function of cycle number for the bare and graphene-wrapped Li4Ti5O12 at different rates from 0.5C to 20C (1C = 175 mAhg1) between 1.0 and 3.0 V. Compared with the bare Li4Ti5O12, the graphene-wrapped electrode has higher discharge capacity and superior rate capability. The capacity difference between the two composites can be obviously found and become larger at higher rates. The enhanced rate capability of [email protected] electrode could be attributed to the reduced polarization and resistance of the electrode because high-conductive graphene networks in the electrode provide more pathways for electron transport [23,24]. As a result, the electrical conductivity of the electrode is enhanced and the polarization of the electrode is correspondingly reduced. Fig. 4(a) shows cyclic voltammetry (CV) of the bare and graphene-wrapped Li4Ti5O12 composites obtained at a scan rate of

291

0.1 mV s1. Both electrodes display a pair of redox peaks at around 1.5 V, corresponding to the partial variation of Ti3+/Ti4+ redox couple [25,26]. The higher and sharper redox peaks signifies faster Li+ insertion/deinsertion in the [email protected] electrode [27]. In addition, the graphene-wrapped Li4Ti5O12 electrode also show smaller potential difference (0.174 V) between two redox peaks than that (0.243 V) of Li4Ti5O12, indicating graphene-wrapping on the surface of Li4Ti5O12 helps to reduce the polarization. Fig. 4(b) shows AC impedance spectra of Li4Ti5O12 and [email protected] electrodes at the stable voltage of 1.55 V. Both EIS curves show a depressed semicircle at the high frequency region and an inclined line at the low frequency region, which reflects the charge-transfer process and the li-ion diffusion process, respectively [28,29]. AC impedance spectra are fitted with the equivalent circuits shown in the inset of Fig. 4(b), where Rs is the solution resistance and Rct was the charge transfer resistance at the particle/electrolyte interface. According to the equivalent circuit, the values of Rct for the Li4Ti5O12 and [email protected] electrodes are calculated to be 34.29 and 21.63 V, indicating a decreased charge transfer resistance for the as-prepared composite after the introduction of graphene nanosheets. It is expected that the formation of some new carbon-oxygen bonds between Li4Ti5O12 and graphene during the calcination process would faciliate electron transport in electrodes [30,31]. The physical properties of coating layer are considered to play a crucial role in the li-ion kinetic behavior in the composites with

Fig. 5. The surface potential maps of (a) Li4Ti5O12 and (b) [email protected] electrodes before cycling; (c) the surface potential map of Au as reference sample; (d) Work functions of Li4Ti5O12 and [email protected] graphene electrodes.

292

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Fig. 6. The energy-level model for explaining the enhanced electron transfer in Li4[email protected] electrodes.

surface modification. The electric properties of materials are characterized by Kelvin probe atomic force microscopy. Fig. 5(a,b) presents the surface potential maps of Li4Ti5O12 and [email protected] electrodes before cycling. According to our prior work [32], the work functions of Li4Ti5O12 and [email protected] are calculated from the surface potential profiles and the results are present in Fig. 5(d). Here, The work functions of the SFM-tip (ftip) is calibrated by Au foil reference sample, whose work function (fAu) is 5.31 eV. It is found that [email protected] possesses a smaller work function (5.04 eV) than that (5.28 eV) of the Li4Ti5O12. The measured work function of Li4Ti5O12 is close to the reported value (5.27 eV) [33]. The smaller work function indicates that the electrons requires less energy to escape from the composite, in turn enhancing the electrochemical performance. These results are consist with EIS measurements. The reduced work function of [email protected] could be explained phenomenologically using the energy-band model, which is schematically illustrated in Fig. 6(a). Due to the disparity in work functions between Li4Ti5O12 and graphene(4.43 eV) [34], electrons transfer occurs from graphene with the smaller work function to Li4Ti5O12 with the larger work function until the two Fermi levels are aligned [35]. As a result, the work function of [email protected] reduces. The smaller work function, the less energy required for electrons to escape from the composites. As shown in Fig. 6(b), more electrons prefers to transfer across the interface between Li4Ti5O12 and graphene, rather than the interface among Li4Ti5O12 particles. As a result, the electron transfer in the electrode is enhance and the polarization is correspondingly reduced. These results are consist with EIS measurements. Considering graphene as a promising anode material, the cells are also tested with an extended lower voltage cutoff at 0.01 V. Fig. 7(a,b) depicts the rate capabilities and cyclic performances of the Li4Ti5O12 and [email protected] electrodes between 0.01 and 3.0 V. Compared to Li4Ti5O12, [email protected] possesses higher

Fig. 7. (a) Rate capabilities and (b) cyclic performances of the Li4Ti5O12 and [email protected] electrodes between 0.01 and 3 V.

specific capacity, improved cyclic performance and better rate capability. For instance, [email protected] electrode showed a reversible capacity of 282.7 mA hg1 after 200 cycles with ca.7.5% capacity loss, while the impressive capacity retention is only ca.83.6% for Li4Ti5O12 electrode. It has been reported that spontaneous Li+ and O2 transfer from the Li4Ti5O12 surface occurs in the LTO/electrolyte interface, resulting in the local formation of disordered-spinel phase such as LiTi5O10.5 [36]. Such disorderedspinel phase has less conductivity, and consequently deteriorate the electrochemical performance of the Li4Ti5O12 electrode. Therefore, it is quite reasonable to assume that the graphenewrapping would not only facilitate electron transfer among Li4Ti5O12 particles but also improve the surface stability by preventing active materials from electrolyte corrosion [37,38]. Fig. 8(a–d) shows the discharge/charge curves of Li4Ti5O12 and [email protected] electrodes as a function of current densities recorded at 1.0-3.0 V and 0.01-3.0 V. Both electrodes exhibit a welldefined flat operational potential plateau at about 1.55 V, which is a typical characteristic of the insertion process of Li+ in spinel Li4Ti5O12 [39]. However, the operational potential plateau becomes shorter and gradually bends down. The longer plateaus of [email protected] indicates a stable biphasic interface in composite [40].The degree of bending of the plateau reflects the polarization of the electrodes. In comparison, [email protected] demonstrates lower polarization and better reaction kinetics, resulting from the enhanced electrical conductivity induced by the grapheme in the composite.

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

293

Fig. 8. The discharge/charge curves of (a) Li4Ti5O12 and (b) [email protected] electrodes at different C rates between 1.03 V; The discharge/charge curves of the (c) Li4Ti5O12 and (d) [email protected] electrodes at different current densities between 0.013 V.

To further investigate the effect of the graphene-wrapping on the Li+ diffusion kinetics of Li4Ti5O12, the impedance spectra under different charge/discharge states for Li4Ti5O12 and [email protected] are continuously measured during the lithium-ion insertion/extraction process. The EIS spetra (Z00 vs. Z0 ) of Li4Ti5O12 and [email protected] are shown in Fig. 9. The chemical diffusion coefficient of Li+ (Dþ Li ) could be calculated based on the relationship between the real axis (Z0 ) and the reciprocal square root(v1/2) in the low frequency region as follow [38] Z 0 ¼ Re þ Rct þ s v v1=2

Dþ Li ¼

2  2RT 2R2 T 2 ¼ pffiffiffi 2n2 F 2 s w AC n4 F 4 s 2w A2 C 2

ð1Þ

ð2Þ

where T, F, R and A are absolute temperature, Faraday’s constant, gas constant and surface area of the electrode respectively. C was molar volume of active material. n was the number of electrons per molecule during oxidation. s w was the Warburg coefficient. Fig. 10 shows lithium diffusion behaviors of Li4Ti5O12 and [email protected] during the lithium-ion insertion/extraction process. Both electordes exhibit similar Li+ diffusion behavior and display a peak around at 1.55 V, which resulted from the two-phase insertion

reaction between Li4Ti5O12 and Li7Ti5O12. In comparison, [email protected] electrode demonstrates larger Li+ diffusion coefficients and smaller difference between insertion and deinsertion peak, indicating that graphene-wrapping would readily facilitate the Li+ diffusion and reduce electrode polarization during the lithium-ion insertion/extraction process. On the other hand, it is found that the diffusion coefficient of Li4Ti5O12 increases up to its maximum (4.3  1012 cm2 s1) as the cell is discharged from 3.0 V to 0.01 V while the diffusion coefficient sharply decreases in the subsequent deinsertion process, which is attributed to the formation of SEI film on Li4Ti5O12 surface. In contrast, the diffusion coefficient decay of [email protected] is much smaller than that of Li4Ti5O12. To get an insight into the mechanism underlying the improved electrochemical performances of Li4Ti5O12 by graphene wrapping, EIS measurements are carried out on Li4Ti5O12 and [email protected] after 200 cycles from 0  C to 30  C, shown in Fig. 11(a,b). The EIS profiles are fit based on the equivalent circuit (present in Fig. 11(c)) and the obtained results reveal that [email protected] demonstrates smaller charge-transfer resistance at the electrolyte/ electrode interface when the operation temperature decreases [41]. A larger Rct generally reflects lower kinetics of the faradic reaction [42]. The charge-transfer resistance (Rct) is strongly associated with the activation energy (4G) of electrode reaction

294

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Fig. 9. The impedance spectra of Li4Ti5O12 and [email protected] electrodes under different charge/discharge states.

and the 4G of bare and graphene-coated Li4Ti5O12 electrodes could be calculated as follows, log Rct ¼ log A þ

DG  R 2:303RT

ð3Þ

where T, R and A are absolute temperature, gas constant and a temperature-independent constant, respectively. Based on the linear relationship between logRct and operation temperature (1000/T) shown in Fig. 11(d), the 4G for Li4Ti5O12 and [email protected] are calculated to be 64.21 KJ mol1 and 52.55 KJ mol1, respectively. Smaller activation energy directly

indicates faster Li ion diffusion [43]. It is expected that the graphene-wrapping layer in composite acts as a catalyst, promoting the charge transfer process at the electrolyte/electrode interface, which could well elucidate the improved Li+ diffusion kinetics in Fig. 11(e). Fig. 12(b,c) presents the surface potential variation across scanned areas of 100nm  100 nm of Li4Ti5O12 and [email protected] discharged to 0.01 V. The testing electrode is consisted of Li4Ti5O12 and [email protected], shown in Fig. 12(a). After discharging to 0.01 V, the cell is disassembled in an argon-filled glove box and the testing electrode is washed several times with

Fig. 10. Lithium ion diffusion coefficients at different charge/discharge states for (a) Li4Ti5O12 and (b) [email protected] electrodes.

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

295

Fig. 11. EIS for (a) Li4Ti5O12 and (b) [email protected] at the full discharge state at different operation temperatures; (c) The equivalent circuit for EIS fitting; (d) Profile of log (Rct) vs. temperature for Li4Ti5O12 and [email protected] electrodes; (e) image figure of improvement of activation energy by graphene coating.

296

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

Fig. 12. (a) The work electrode consisting of Li4Ti5O12 and [email protected]; The surface potential maps of (b) Li4Ti5O12 and (c) [email protected] electrodes; (d) Work functions of Li4Ti5O12 and [email protected] graphene electrodes.

dimethyl carbonate and ethanol, subsequently followed by an overnight drying in a vacuum. The calculated work function (4.09 eV) of Li4Ti5O12 is larger than that (3.93 eV) of [email protected] It is well established that the SEI film is formed onto the anode electrode surface due to electrolyte decomposition when the electrode was discharged below 1.0 V vs Li/Li+. The smaller work function indicates the less energy required for electrons to escape from the composites. The enhanced electron transfer of [email protected] could be explained based on phenomenological resistance model. Fig. 13 presents the possible electron-transfer behavior in the [email protected] electrode and the corresponding equivalent electrical circuit between Li4Ti5O12 particles (Fig. 13b), where RLTO, RG, RSEI and RPVDF present Li4Ti5O12 resistance, graphene resistance, SEI-film resistance and PVDF resistance, respectively. According to the parallel circuit model, the loss of electrical contact among Li4Ti5O12 particles would be reduced, in which the total resistance is strongly dependent upon the high-conductivity graphene. As a result, the electron-transfer in [email protected] electrode is enhanced and the electrochemical performances are correspondingly improved.

The heterojunction interface between Li4Ti5O12 and graphene seems to make a positive impact on the improved performance. The disparity in work functions between Li4Ti5O12 and graphene results in electrons transfer from graphene to Li4Ti5O12 when they are contacted [31]. A very small transfer of charge builds up a surface charge and a corresponding electric field(E) between them, shown in Fig. 14(a). Such electric field could facilitate Li-ion diffusion from graphene side to Li4Ti5O12 side, and electron transfer from Li4Ti5O12 side to graphene side across heterojunction interfaces, shown in Fig. 14(b). During the lithium insertion process, more electrons in Li4Ti5O12 tend to transfer into graphene rather than Li4Ti5O12 surface under the electric field(E). It is expected that graphene are more responsible for the reduction of carbonate electrolyte and the formation of solid-electrolyte interface(SEI) on Li4Ti5O12 is corresponding suppressed. Therefore, [email protected] might have better lithium-ion diffusion at the interface of Li4Ti5O12/electrolyte. On the another hand, graphene wrapped on Li4Ti5O12 could provide additional intercalation sites and diffusion channels for lithium-ion diffusion channels in Li4Ti5O12. The reasonable analyses are consistent with the obtained EIS and workfunction measurements.

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

297

subsequently wrapped by graphene through electrostatic interactions. In comparison, [email protected] exhibited superior electrochemical performances. Analysis from the electrochemical measurements reveals that the improved electrochemical performances of [email protected] should be attributed to lower charge-transfer resistances, larger lithium-ion diffusion coefficient and lower activation energy. KPFM measurements indicate that the electrons transfer at Li4Ti5O12/graphene heterojunction interface reduces the localized work function of the composites, decreases energy required for electrons to escape and consequently results in the improved electrochemical performances. Acknowledgements This work was supported by a grant from Natural Science Foundation of Fujian Province (No.2013J01007), Key Project of Department of Science & Technology of Fujian Province (No.2014H0020), Program for New Century Excellent Talents in University of Fujian Province (No.JA14069) and Project A of Education Bureau of Fujian Province (No.JA12067). References

Fig. 13. (a) Possible electron-transfer behavior in the [email protected] electrode; (b) the equivalent electrical circuit between Li4Ti5O12 particles.

Fig. 14. A built electric field(E) between Li4Ti5O12 and [email protected] graphene; (b) possible transfer behavior of lithium-ions and electrons under electric field.

4. Conclusions Flower-like Li4Ti5O12 hollow microspheres consisting of nanosheets were prepared by hydrothermal technique, and

[1] J. Jaguemont, L. Boulon, Y. Dubé, A comprehensive review of lithium-ion batteries used in hybrid and electric vehicles at cold temperatures, Appl. Energy 164 (2016) 99. [2] J. Lu, Z.H. Chen, Z.F. Ma, F. Pan, A.L. Curtiss, K. Amine, The role of nanotechnology in the development of battery materials for electric vehicles, Nature Nanotechnology. 11 (2016) 1031. [3] L.H. Saw, Y.H. Ye, A.A.O. Tay, Integration issues of lithium-ion battery into electric vehicles battery pack, J. Clean. Prod. 113 (2016) 1032. [4] B.T. Zhao, R. Ran, M.L. Liu, Z.P. Shao, A comprehensive review of Li4Ti5O12-based electrodes for lithium-ion batteries: The latest advancements and future perspectives, Mater. Sci. Engin. R. 98 (2015) 1. [5] X.C. Sun, P.V. Radovanovic, B. Cui, Advances in spinel Li4Ti5O12 anode materials for lithium-ion batteries, New J.Chem. 39 (2015) 38. [6] Z. Yuan, P. Tan, C.R. Ma, J.H. Yang, S.Y. Ma, S.Y. Zheng, Challenges of spinel Li4Ti5O 12 for lithium-ion battery industrial applications, Adv. Energy Mater. (2017), doi:http://dx.doi.org/10.1002/aenm.201601625. [7] X. Bai, T. Li, Z.Y. Dang, Y.X. Qi, N. Lun, Y.J. Bai, Ionic conductor of Li2SiO3 as an effective dual-functional modifier to optimize the electrochemical performance of Li4Ti5O12 for high-performance Li-ion batteries, ACS Appl. Mater. Interfaces 9 (2017) 1426. [8] X.Y. Feng, H.L. Zou, H.F. Xiang, X. Guo, T.P. Zhou, Y.C. Wu, W. Xu, P.F. Yan, C.M. Wang, J.G. Zhang, Y. Yu, Ultrathin Li4Ti5O12 nanosheets as anode materials for lithium and sodium storage, ACS Appl. Mater. Interfaces 8 (2016) 16718. [9] X. Bai, W. Li, A.J. Wei, X.H. Li, L.H. Zhang, Z.F. Liu, Preparation and electrochemical properties of Mg2+ and F co-doped Li4Ti5O12 anode material for use in the lithium-ion batteries, Electrochim. Acta. 222 (2016) 1045. [10] J.W. Zhang, Y.R. Cai, J. Wu, J.M. Yao, Graphene oxide-confined synthesis of Li4Ti5O12 microspheres as high-performance anodes for lithium ion batteries, Electrochim. Acta. 165 (2015) 422. [11] A. Nugroho, K.Y. Chung, J. Kim, A facile supercritical alcohol route for synthesizing carbon coated hierarchically mesoporous Li4Ti5O12 microspheres, J. Phys. Chem. C 118 (2014) 183. [12] P. Roy, S.K. Srivastava, Nanostructured anode materials for lithium ion batteries, J. Mater. Chem. A 3 (2014) 2454. [13] R. Xue, J.W. Yan, L. Jiang, B.L. Yi, Fabrication of lithium titanate/graphene composites with high rate capability as electrode materials for hybrid electrochemical supercapacitors, Mater. Chem. Phys. 160 (2015) 375. [14] L. Cheng, J. Yan, G.N. Zhu, J.Y. Luo, C.X. Wang, Y.Y. Xia, General synthesis of carbon-coated nanostructure Li4Ti5O12 as a high rate electrode material for Liionintercalation, J. Mater. Chem. 20 (2010) 595. [15] H.P. Liu, G.W. Wen, S.F. Bi, C.Y. Wang, J.M. Hao, P. Gao, High rate cycling performance of nanosized Li4Ti5O12/graphene composites for lithium ion batteries, Electrochim.Acta. 192 (2016) 38. [16] F.L. Zhang, B. Xu, G.P. Cao, M. Chu, N. Biao, G. Wei, Y.S. Yang, Nano/micro structured porous Li4Ti5O12 synthesized by a polyethylene glycol assisted hydrothermal method for high rate lithium-ion batteries, RSC Adv. 4 (2015) 53981. [17] L.K. Tang, Y.B. He, C. Wang, S. Wang, M. Wagemaker, B.H. Li, Q.H. Yang, F.Y. Kang, High-density microporous Li4Ti5O12 microbars with superior rate performance for lithium-ion batteries, Adv. Sci. (2016), doi:http://dx.doi.org/ 10.1002/advs.201600311. [18] J. Cheng, R.C. Che, C.Y. Liang, J.W. Liu, M. Wang, J.J. Xu, Hierarchical hollow Li4Ti5O12 urchin-like microspheres with ultra-high specific surface area for high rate lithium ion batteries, Nano Research. 1043 (2014) 7.

298

Z. Lin et al. / Electrochimica Acta 254 (2017) 287–298

[19] B.T. Zhao, X. Deng, R. Ran, M.L. Liu, Z.P. Shao, Facile synthesis of a 3D nanoarchitectured Li4Ti5O12 electrode for ultrafast energy storage, Adv. Energy Mater. 6 (2016) 1500924. [20] H. Ge, T.T. Hao, B. Zhang, L. Chen, L.X. Cui, X.M. Song, Nanoparticlesconstructed spinel Li4Ti5O12 with extra surface lithium storage capability towards advanced lithium-ion batteries, Electrochim Acta. 211 (2016) 119. [21] L.J. Cao, Y.N. Hui, H.B. Ouyang, J.F. Huang, Z.W. Xu, J.Y. Li, W.Z. Zhang, S.M. Chai, S.W. Guo, Li4Ti5O12 hollow mesoporous microspheres assembled from nanoparticles for high rate lithiumion battery anodes, RSC Adv. 5 (2015) 35643. [22] N.S. Xu, X.Z. Sun, X. Zhang, K. Wang, Y.W. Ma, A two-step method for preparing Li4Ti5O12-graphene as an anode material for lithium-ion hybrid capacitors, RSC Adv. 5 (2015) 94361. [23] Y. Shi, L. Wen, F. Li, H.M. Cheng, Nanosized Li4Ti5O12/graphene hybrid materials with low polarization for high rate lithium ion batteries, J Power Sources. 196 (2011) 8610. [24] J.P. Zhu, R. Duan, Y.Y. Zhang, J. Zhu, A facial solvothermal reduction route for the production of Li4Ti5O12/graphene composites with enhanced electrochemical performance, Ceram. Int. 42 (2016) 334. [25] M. Kitta, T. Akita, Y. Maeda, M. Kohyama, Study of surface reaction of spinel Li4Ti5O12 during the first lithium insertion and extraction processes using atomic force microscopy and analytical transmission electron microscopy, Langmuir. 28 (2012) 12384. [26] K. Zaghib, M. Simoneau, M. Armand, Electrochemical study of Li4Ti5O12 as negative electrode for Li-ion polymer rechargeable batteries, J. Power Sources. 81 (1999) 300. [27] Z.W. Xie, X. Li, W. Li, M.Z. Chen, M.Z. Qu, Graphene oxide/lithium titanate composite with binder-free as high capacity anode material for lithium-ion batteries, J. Power Source. 273 (2015) 754. [28] Z. Ding, L. Zhao, L. Suo, Y. Jiao, S. Meng, Y.-S. Hu, Z. Wang, L. Chen, Towards understanding the effects of carbon and nitrogen-doped carbon coating on the electrochemical performance of Li4Ti5O12 in lithium ion batteries: a combined experimental and theoretical study, Phys. Chem. Chem. Phys. 13 (2011) 15127. [29] H. Zhang, Y. Liu, T. Wang, Y. Yang, S.J. Shi, G. Yang, Li2ZrO3-coated Li4Ti5O12 with nanoscale interface for high performance lithium-ion batteries, Appl. Surf. Sci. 368 (2016) 56. [30] Y. Ding, C.W. Xiao, X.P. Gao, Insight into effects of graphene in Li4Ti5O12/carbon composite with high rate capability as anode materials for lithium ion batteries, Electrochim. Acta. 102 (2013) 282.

[31] H.P. Liu, G.W. Wen, S.F. Bi, C.Y. Wang, J.M. Hao, P. Gao, High rate cycling performance of nanosized Li4Ti5O12/graphene composites for lithium ion batteries, Electrochim. Acta. 192 (2016) 38. [32] L.C. Chen, Y.M. Yang, Z.S. Wang, J.Y. Zhang, Q.L. Su, Y. Chen, W. Chen, Y.B. Lin, Z. G. Huang, Enhanced electrochemical performances and thermal stability of LiNi1/3Co1/3Mn1/3O2 by surface modification with YF3, J. Alloy Compd. 711 (2017) 462. [33] B.J. Morgan, J. Carrasco, G. Teobaldi, Variation in surface energy and reduction drive of a metal oxide lithium-ion anode with stoichiometry: a DFT study of lithium titanate spinel surfaces, J. Mater. Chem. A. 4 (2016) 17180. [34] X.W. Wang, G.Z. Sun, P. Routh, D.-H. Kim, W. Huang, P. Chen, Heteroatomdoped graphene materials: syntheses, properties and applications, Chem. Soc. Rev. 43 (2014) 7067. [35] B. Bhushan, Scanning probe microscopy in nanoscience and nanotechnology, Springer Heidelberg, Dordrecht London New York, 2017. [36] J. Liu, C.Q. Du, Z.Y. Tang, M. Yang, X.H. Zhang, In situ nickel/carbon coated lithium titanium oxide anode material with improved electrochemical properties, Electrochim. Acta. 143 (2014) 56. [37] M. Kitta, T. Akita, M. Kohyama, Spontaneous li-ion transfer from spinel Li4Ti5O12 Surfaces: deterioration at Li4Ti5O12/electrolyte interfaces stored at room temperature, J. Electrochem. Soc. 162 (2015) A1272. [38] Z.Y. Lin, W.B. Zhu, Z.S. Wang, Y.M. Yang, Y.B. Lin, Z.G. Huang, Synthesis of carbon-coated Li4Ti5O12 nanosheets as anode materials for high-performance lithium-ion batteries, Journal of Alloys and Compounds. 687 (2016) 232. [39] G.E. Luo, J.R. He, X.J. Song, X.Y. Huang, X.Y. Yu, Y.P. Fang, D.Y. Chen, Bamboo carbon assisted sol-gel synthesis of Li4Ti5O12 anode material with enhanced electrochemical ctivity for lithium ion battery, J. Alloys Compd. 621 (2015) 268. [40] J. Liu, C.Q. Du, Z.Y. Tang, M. Yang, X.H. Zhang, In situ nickel/carbon coated lithium titanium oxide anode material with improved electrochemical properties, Electrochim. Acta. 143 (2014) 56. [41] T. Yuan, X. Yu, R. Cai, Y. k. Zhou, Z.p. Shao, Synthesis of pristine and carboncoated Li4Ti5O12 and their low-temperature electrochemical performance, J. Power Sources. 195 (2010) 4997. [42] Y.B. Lin, Y. Lin, T. Zhou, G.Y. Zhao, Y.D. Huang, Z.G. Huang, Enhanced electrochemical performances of LiFePO4/C by surface modification with Sn nanoparticles, J Power Sources. 226 (2013) 20. [43] M. Okubo, Y. Tanaka, H.S. Zhou, I. Honma, Determination of activation energy for li ion diffusion in electrodes, J. Phys. Chem. B. 113 (2009) 2840.