The Li4Ti5O12/graphene composite presents excellent rate capability and low-temperature performance. Even at 120 C, it still delivers a quite high capacity of about 136 mAh g−1. When the charge–discharge tests are performed at −10 ◦C and −20 ◦C, its specific capacities are as high as 149 and 102 mAh g−1, respectively. In addition, the full-cells using LiNi1/3Co1/3Mn1/3O2 as cathode material exhibit good rate capability.

1. Introduction

Lithium-ion batteries (LIBs) have been regarded as promising power sources for electric vehicles (EVs), owing to their advantages on energy density and lifetime. But the state-of-the-art LIBs based on the graphite anode can hardly meet the requirements of EVs on power density and safety characteristics [1]. Especially, limitation of lithium diffusion kinetics in the graphite crystals results in quite a long time to fully charge. Recently, novel candidates of graphite have attracted great interests, among which lithium titanate (Li4Ti5O12, LTO) is the most promising for applications in the large-scale LIBs [2–8]. It exhibits excellent safety characteristics and long lifetime [7,8]. Moreover, nanosized Li4Ti5O12 particles have short paths for lithium ion and electron transport and help to achieve fast charging for the LIBs [7]. However, the poor electronic conductivity of Li4Ti5O12 restricts its rate performance [9,10]. Many approaches have been developed to solve this problem, including morphology tailoring and nanostructuring, ion doping, surface modification and mixing with some extraordinary conductive components [11–21]. Graphene, with excellent electronic conductivity, high surface area, outstanding thermal properties and mechanical strength, is considered as an ideal conductive additive to the nanostructured composites that used in LIBs and other electrochemical devices [22,23]. Hence the composites containing Li4Ti5O12 and graphene have been widely investigated as high rate anode materials for lithium-ion batteries [24–29]. For example, Zhu et al. have fabricated grapheneembeddedLi4Ti5O12 nanofibers by electrospinningdeposition[24]; Shen et al. and Tang et al. both prepared graphene–Li4Ti5O12 hybrid nanostructures using a two-step hydrothermal reaction [25,26]; and we have prepared Li4Ti5O12/graphene composite in a sol–gel method [27]. Ding et al. clarified the positive effects of graphene in Li4Ti5O12/carbon composites as anode materials recently [29]. Although Li4Ti5O12/graphene composites mentioned above exhibited good electrochemical performance, it still remains a great challenge to develop a facile and economical preparation route to achieve high performance Li4Ti5O12/graphene composites. Solid state reaction is a commonly used method to prepare electrode materials in the battery industry. Nevertheless, to the best of our knowledge, there was no report that prepared homogeneous Li4Ti5O12/graphene composites by an in situ solid state reaction. Here in our investigation, inhomogeneity of Li4Ti5O12/graphene composites results from facile stacking of graphene sheets and agglomeration of the Li4Ti5O12 nanoparticles in a typical solid state reaction. By a carbon pre-coating process, we successfully overcomed this issue and prepared the homogeneous Li4Ti5O12/graphene composite in an in situ solid state reaction. The structure and morphology of the Li4Ti5O12/graphene composites were investigated, and their cell performance was also evaluated, including rate capability and low-temperature performance.

2. Experimental

2.1. Materials preparation

The graphene sheets were prepared via a thermal exfoliation route, including graphite oxidation through modified Hummer method, followed by rapid thermal expansion of graphene oxide at 1050 ◦C in nitrogen atmosphere [30]. Anatase TiO2 and glucose with a weight ratio of 4:1 were mixed in ethanol–water compounds (10:1 in volume) and stirred for 2 h, then drying for 10 h under air-circulating oven at 100 ◦C. The mixture was heated at 600 ◦C for 5 h under N2 atmosphere to obtain the carbon coated TiO2. The carbon content in the coated TiO2 was about 6 wt.%. Graphene, Li2CO3 and carbon-coated TiO2 were dispersed in hexamethylene, and then mixed by balling milling, which was performed in a planetary ball mill (Nanjing) under atmosphere at rotational speed of 200 rpm for 4 h. The as-obtained slurry was dried and further calcined at 800 ◦C for 12 h in N2 atmosphere to obtain the carbon-coated Li4Ti5O12/graphene composites (denoted as C-LTO/graphene). In the composite, the total content of carbon including coated carbon and graphene sheets is about 10%. For comparison, pristine Li4Ti5O12 (LTO) and Li4Ti5O12/graphene composites without carbon coating (LTO/graphene) were prepared in a similar way.

2.2. Materials characterization

The crystal structure of the products was identified by Xray diffraction (XRD) measurements (D/max 2500V) using Cu K ( = 1.5406A)˚ radiation in the 2 range of 10–70◦. The morphology of the products was observed by scanning electron microscopy (SEM, JEOL JSM-6390LA) and transmission electron microscopy (TEM, JEM-2100F). 2

2.3. Electrochemical measurements

The electrochemical performance of the products was evaluated in coin-type cells. In order to make an electrode laminate, a uniform slurry containing 84 wt.% active material, 8 wt.% acetylene black and 8 wt.% polyvinylidene fluoride (PVDF) dispersed in N-methyl-2-pyrrolidinone (NMP) was cast onto an copper current collector. After vacuum drying at 70 ◦C, the laminate was punched into discs (˚ 14 mm) for assembling the cells. The mass loading in the electrode was controlled at about 5 mg cm−2. In the half-cells, various LTO materials were used as working electrode and high-purity lithium metal as counter electrode. In the full-cells, the LiNi1/3Co1/3Mn1/3O2 positive electrode prepared in a similar way (just the current collector was replaced with aluminium foil) and the C-LTO/graphene negative electrode had the same mass loading, and the cell capacity was calculated on the mass of LiNi1/3Co1/3Mn1/3O2. The electrolyte was a solution of 1 M LiPF6 in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1, w/w). The separator is Celgard 2400 microporous polypropylene membrane. The cell performance of samples was evaluated on a multi-channel battery cycler (Neware BTS2300). Galvanostatic charge–discharge tests were performed under different current rates, where 1 C is corresponding to the current density of 150 mAg−1. For the half-cells and the full-cells, the cutoff voltages were set as 2.5–1.0V and 3.0–1.0V, respectively. The temperature for the low-temperature performance measurements was controlled by a low-temperature chamber (Shanghai Yiheng Instruments Co., Ltd.). The AC impedance spectrum of the cells was measured by a CHI 604D electrochemical workstation (Shanghai Chenhua Instruments Co., Ltd.)in the frequency range from 10 mHz to 100 kHz with a potential perturbation at 5 mV.

3. Results and discussion

The crystal structures of Li4Ti5O12 and its graphene composites are identified as shown in Fig. 1. The diffraction peaks of the pristine LTO, LTO/graphene composite and C-LTO/graphene composite are similar and conform to JCPDS card No.49-0207 in accordance with the spinel Li4Ti5O12 phase with Fd3m space group. The narrow diffraction peaks indicate that Li4Ti5O12 in the three samples are all highly crystalline. Similar as the XRD patterns of most Li4Ti5O12/C composites [6,31,32], no any diffraction according to graphitic structure was detected, mainly due to the amorphous structure of coated carbon and graphene sheets. All the results indicate that the addition of graphene sheets and carbon-coating have no impact on the crystal structure and crystallinity of spinel Li4Ti5O12 in the solid state reaction. Fig. 2 shows SEM images of LTO, LTO/graphene, C-LTO/graphene and TEM image of C-LTO/graphene composite, respectively. As shown in Fig. 2a, the Li4Ti5O12 primary particles are 200–800 nm in size and they are apt to aggregate together. The image of LTO/graphene (Fig. 2b) shows that the Li4Ti5O12 particles are almost separated with graphene sheets. Moreover, particle size and aggregation degree of Li4Ti5O12 in the LTO/graphene composite are similar to the pristine Li4Ti5O12. Seemingly, it is difficult for bare LTO particles to stably anchor on the graphene sheets. Especially, aggregation of the Li4Ti5O12 nanoparticles and further agglomeration result in inhomogeneity of Li4Ti5O12/graphene composites and even phase separation. From the images of C-LTO/graphene composite (Fig. 2c and d), it can be seen that the Li4Ti5O12 particles are quite uniformly dispersed among a three-dimensional network built by the graphene sheets, even though a few agglomerates are still inevitable in the high-temperature solid state reaction. The graphene network gives a conductive connection between the Li4Ti5O12 particles. The size of most Li4Ti5O12 particles is less than 100 nm (Fig. 2d). Furthermore, as revealed by the TEM image in Fig. 2d inset, the coated carbon layer has a thickness of less 10 nm. Basically, it can be easily concluded that the homogeneous......

Details, please click:http://web.mit.edu/~elsao/www/Articles/2013Guo%20Solid-state%20synthesis%20and%20electrochemical%20performance%20of%20Li4Ti5O12-graphene%20composit%20for%20lithium-iion%20batteries.pdf