Graphene wrapped ordered LiNi0.5Mn1.5O4 nanorods as promising cathode material for lithium-ion batteries

LiNi0.5Mn1.5O4 nanorods wrapped with graphene nanosheets have been prepared and investigated as high energy and high power cathode material for lithium-ion batteries. The structural characterization by X-ray diffraction, Raman spectroscopy, and Fourier transform infrared spectroscopy indicates the LiNi0.5Mn1.5O4 nanorods prepared from β-MnO2 nanowires have ordered spinel structure with P4332 space group. The morphological characterization by scanning electron microscopy and transmission electron microscopy reveals that the LiNi0.5Mn1.5O4 nanorods of 100–200 nm in diameter are well dispersed and wrapped in the graphene nanosheets for the composite. Benefiting from the highly conductive matrix provided by graphene nanosheets and one-dimensional nanostructure of the ordered spinel, the composite electrode exhibits superior rate capability and cycling stability. As a result, the LiNi0.5Mn1.5O4-graphene composite electrode delivers reversible capacities of 127.6 and 80.8 mAh g−1 at 0.1 and 10 C, respectively, and shows 94% capacity retention after 200 cycles at 1 C, greatly outperforming the bare LiNi0.5Mn1.5O4 nanorod cathode. The outstanding performance of the LiNi0.5Mn1.5O4-graphene composite makes it promising as cathode material for developing high energy and high power lithium-ion batteries.

with Fd-3m symmetry, while Mn 4+ and Ni 2+ ions order in stoichiometric LiNi 0.5 Mn 1.5 O 4 resulting in the P4 3 32 symmetry 7,8 . The disordered spinel shows higher electronic conductivity due to the presence of Mn 3+ /Mn 4+ redox couple and Ni/Mn disordering 9 . However, the nonstoichiometry in disordered spinel often induces impurities such as NiO and Li x Ni y O, and the 4 V (vs. Li/Li + ) Mn 3+ /Mn 4+ redox couple, thus reducing the specific capacity and energy 10 . It is still a challenge to achieve both high energy and high power densities for the ordered spinel due to its poor electronic conductivity. Moreover, the high working voltage at about 4.7 V (vs. Li/Li + ) induces side reactions at the electrode/electrolyte interface, resulting in continuous capacity fading during the cycling 11 .
To circumvent the drawbacks of the high voltage spinels, two strategies are often employed to improve the electrochemical performance. One approach is to construct nanostructures for the spinel, which can drastically shorten the transport distance for both electrons and lithium ions, resulting in greatly improved rate capability [12][13][14][15] . However, the increased surface area associated with nanostructuring will aggravate the side reactions at high voltage and deteriorate the capacity fading during cycling. Another approach people often used is to modify the spinel surface by coating a thin protective layer, which can greatly improve the interface stability with enhanced cycling performance [16][17][18][19][20] . However, the coating materials, most of which are metal oxides and metal fluorides, only function as protective layers for the spinel without improvement in electrical conductivity. Therefore, new strategies that can combine the advantages of the above mentioned two approaches need to be developed to further improve the electrochemical performance of the high voltage spinels.
Herein, we developed a facile method to prepare graphene nanosheets wrapped ordered LiNi 0.5 Mn 1.5 O 4 nanorods as high energy and high power cathode material for lithium-ion batteries (Fig. 1). In the hybrid electrode design, the one-dimensional nanostructure of LiNi 0.5 Mn 1.5 O 4 enables fast lithium ion transport while the graphene wrapping suppresses the side reactions at high voltage and further improves the electron transfer. It has been demonstrated that the graphene or grahene oxide nanosheets incorporation can greatly improve the rate capability and cycling stability for the disordered spinels 21,22 . Consequently, the ordered LiNi 0.5 Mn 1.5 O 4 nanorods-graphene composite cathode exhibited greatly improved cycling performance and rate performance compared to the bare ordered LiNi 0.5 Mn 1.5 O 4 nanorods. The promising results indicate the great potential of developing high energy and high power lithium-ion batteries by utilizing the graphene nanosheets wrapped ordered LiNi 0.5 Mn 1.5 O 4 nanorods.

Results
The hydrothermally prepared β -MnO 2 nanowires were used as the template and the XRD pattern shows no trace of impurity (Fig. S1, Supporting Information). Figure 2 shows the XRD patterns of the pristine graphene nanosheets, the as-synthesized LiNi 0.5 Mn 1.5 O 4 nanorods, and the LiNi 0.5 Mn 1.5 O 4 -graphene composite. The XRD pattern of the graphene shows a small hump at about 26°, which can be attributed to the (002) reflection of graphite. The as-synthesized LiNi 0.5 Mn 1.5 O 4 nanorods and the LiNi 0.5 Mn 1.5 O 4 -graphene composite show similar XRD patterns, which can be indexed to the cubic spinel structure with space group P4 3 32 (JCPDS No. 80-2184). No impurity peaks from NiO or Li x Ni y O can be detected, indicating the existence of pure spinel phase. Rietveld refinement gives a lattice parameter of a = 8.169 Å, which agrees well with reported value for the ordered LiNi 0.5 Mn 1.5 O 4 12,13 . No diffraction peaks of graphene can be observed from the XRD pattern of the LiNi 0.5 Mn 1.5 O 4 -graphene composite, which is probably due to the strong diffraction peaks from the highly crystalline LiNi 0.5 Mn 1.5 O 4 and the nanoscale size feature of low content graphene. However, the superstructure peaks, which are characteristics of Ni and Mn ordering, cannot be resolved from the XRD patterns of the spinels because of their low intensities 23 . Therefore, further structural investigation by Raman and FTIR are required to confirm the P4 3     The Raman spectrum of the LiNi 0.5 Mn 1.5 O 4 nanorods in the frequency range between 300 to 700 cm −1 was enlarged and shown in Fig. 3b, revealing six Raman bands. The strong band around 638 cm −1 is assigned to the symmetric Mn-O stretching mode of MnO 6 octahedral (A 1g ), while the two bands at 407 and 498 cm −1 are associated with the Ni 2+ -O stretching mode in the structure 25 . The peak near 580-620 cm −1 is considered as T2g (3) of the spinel compound, and the split of T2g (3) is the strong evidence of the ordered spinel due to its low symmetry (P4 3 32). In the frequency range between 1000 and 3700 cm −1 , the Raman spectrum of the LiNi 0.5 Mn 1.5 O 4 -graphene composite show similar Raman features as the pristine graphene nanosheets, revealing the characteristic D band, G band, and 2D band of graphene.
To further confirm the structural symmetry of the spinel in the present work, FTIR analysis was carried out on different spinel powder samples. For comparison, disordered LiNi 0.5 Mn 1.5 O 4 powders were prepared by a solid state synthesis according to the literature 26 . Figure 4a 28 . Previous studies indicate that the redox peak splitting at about 4.7 V (vs. Li/Li + ) for the high voltage Li x Ni 0.5 Mn 1.5 O 4 (0 < x < 1) is probably due to two separate redox couples of Ni 2+ /Ni 3+ and Ni 3+ /Ni 4+ and/or Li/vacancy ordering at x = 0.5 29 . However, such peak splitting is not obvious for the ordered spinel as only one pair of redox peaks are observed in the typical CV curve. As discussed in our previous work on disordered LiNi 0.5 Mn 1.5 O 4-δ thin films 30 , the ordering arrangement of Ni and Mn in the ordered spinel is probably not commensurate with the preferred Li/vacancy ordering at x = 0.5 so that the Ni/Mn ordering suppresses Li/vacancy ordering and redox peak splitting at about 4.7 V (vs. Li/Li + ). In comparison with the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode, the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode shows much smaller peak separation between the cationic peak and anodic peak in the CV curve, indicating the electrode polarization can be greatly reduced by incorporating the graphene nanosheets into LiNi 0.5 Mn 1.5 O 4 nanorods.     31 . The wrapping with graphene could greatly suppress the SEI layer formation at high voltage, thus improving the initial coulombic efficiency of the composite electrode. Figure 8c compares the cycle performance of the two electrodes, revealing greatly improved cycling stability for the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode. After 200 cycles at 0.1 C rate, LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode can still deliver a reversible capacity of about 115 mAh g −1 , retaining 94% of its initial reversible capacity. In comparison, the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode only retained 82% of its initial reversible capacity. The capacity fading of the high voltage spinel during cycling is mainly contributed by the structural deterioration induced by Mn 3+ ion dissolution and internal resistance increase induced by the side reactions at the electrode surface at high voltage 32,33 . For the ordered spinel, Mn 3+ ion dissolution may not be the major reason that causes the capacity fading since there are negligible Mn 3+ ions in ordered spinel due to its nearly perfect stoichiometry. The side reactions, including SEI layer formation, could be more detrimental to the cycle performance because the increased polarization induced by the increasing resistance will lead to less reversible capacity. As shown in Fig. 8a, the voltage difference between charge and discharge keeps increasing with the cycling test, revealing a obvious cell polarization growth for the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode. By contrast, the polarization growth for the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode is greatly mitigated, which can be attributed to the graphene protection, suppressing the side reactions at the electrode surface. Figure 8d compares the rate capability of the two electrodes by plotting the specific capacity as a function of cycle number  Fig. S3 (Supporting Information). It is obvious that the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode possesses much better rate capability as it can retain more reversible capacity as the discharge rate increases. Even at 10 C rate, the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode can still deliver a reversible capacity of about 80.8 mAh g −1 , which is much larger than that of the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode (49.2 mAh g −1 ). When the current rate was set back to 0.1 C, the charge and discharge capacities of LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode recover to the original values, indicating that large current density and rapid lithiation/delithiation did not cause any permanent damage to the crystal structure. However, after the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode experienced the high current rate like 10 C, its reversible capacity didn't fully recover to the initial value when the current rate was changed back to 0.1 C. The superior rate performance of the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode can be attributed to the improved electron transport provided by the graphene conductive matrix. As confirmed by the EIS measurements, the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode shows much smaller charge transfer resistance compared to the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode, indicating the graphene wrapping is beneficial to fast electrode kinetics (Fig. S4, Supporting Information).

Discussion
Although the conductivity of the ordered LiNi 0.5 Mn 1.5 O 4 is relatively low, the ordered LiNi 0.5 Mn 1.5 O 4 nanorod-graphene composite developed in this work exhibited excellent rate capacity as well as good cycling stability. The rate performance of the present ordered LiNi 0.5 Mn 1.5 O 4 -graphene composite is even better than those of the previously reported disordered LiNi 0.5 Mn 1.5 O 4 -grahene composite and disordered LiNi 0.5 Mn 1.5 O 4 -grahene oxide composite 21,22 , which can be attributed to its unique hybrid electrode design. First, the one-dimensional nanostructure of ordered LiNi 0.5 Mn 1.5 O 4 provides short solid diffusion length for lithium ions. This, along with fast electron transport supplied by the highly conductive graphene matrix, endows the hybrid electrode with fast lithiation and delithiation capability. Second, the ordered LiNi 0.5 Mn 1.5 O 4 nanorods with P4 3 32 symmetry contain negligible Mn 3+ ions, which minimizes Mn disproportionative dissolution and Jahn-Teller structural distortion, resulting in good structural stability during cycling. Last, the graphene wrapping effectively modify the surface of LiNi 0.5 Mn 1.5 O 4 , which suppresses the side reactions at the electrode/electrolyte interface, resulting in a slow cell polarization growth and excellent cycling stability. Unlike disordered LiNi 0.5 Mn 1.5 O 4 , the ordered LiNi 0.5 Mn 1.5 O 4 do not have the 4 V voltage plateau, thus making the LiNi 0.5 Mn 1.5 O 4 -grpahene composite more promising for application in high energy and high power lithium-ion batteries.
In summary, a smart hybrid cathode material featuring graphene nanosheets wrapped one-dimensional ordered LiNi 0.5 Mn 1.5 O 4 nanorods has been developed by a facile method combining oxide template synthesis and chemical mixing. The structural characterization confirmed the as-prepared LiNi 0.5 Mn 1.5 O 4 nanorods possess a ordered spinel structure with P4 3 32 symmetry. The LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode exhibited excellent rate capability, delivering reversible capacities of 122.4 and 81.2 mAh g −1 at 0.1 and 10 C, respectively. In addition to the fast charge and discharge capability, the LiNi 0.5 Mn 1.5 O 4 -graphene composite electrode also revealed promising cycling stability, with 94% capacity retained after 200 cycles. In comparison with the bare LiNi 0.5 Mn 1.5 O 4 nanorod electrode, the composite electrode showed greatly improved electrochemical performance, which can be ascribed to the smart hybrid electrode design, enabling both fast charge transport and good interface stability. The superior electrochemical performance and the facile synthesis procedure make the present LiNi 0.5 Mn 1.5 O 4 -graphene composite promising as cathode for high energy lithium-ion batteries.

Methods
Preparation β-MnO 2 nanowires and ordered LiNi 0.5 Mn 1.5 O 4 nanorods. β -MnO 2 nanowires were prepared by a modified hydrothermal method according to the literature 34 . In a typical synthesis, 10 g KNO 3 were added into a solution containing 20 ml 50 wt% MnNO 3 and 20 ml deionized water to get a supersaturated system. Then the well mixed solution was transferred into a 50 mL Teflon-lined stainless steel autoclave and heated at 140 °C for 14 h. After cooling down to room temperature, the precipitated product was filtered and washed sequentially with deionized water and ethanol for three times. The obtained β -MnO 2 nanowires were then dried at 80 °C for 6 h in air for further usage. The ordered spinel LiNi 0.5 Mn 1.5 O 4 nanorods were then synthesized by using the β -MnO 2 nanowires as the template. Briefly, stoichiometric amounts of Ni(CH 3 COO) 2 , LiOH·H 2 O, and β -MnO 2 nanowires were homogeneously dispersed in high purity ethanol and stirred for 24 h. The obtained precursor was dried at 60 °C for 3 h and ground in a mortar for 1 h. After that, the mixture was first preheated at 300 °C for 5 h and then calcined at 700 °C for 10 h in air to obtain the ordered LiNi 0.5 Mn 1.5 O 4 nanorods.
Preparation of LiNi 0.5 Mn 1.5 O 4 -graphene composite. The grapnehe nanosheets were prepared from graphite powder in a two-step process, involving the oxidation and/or exfoliation of graphite to graphite oxide by Hummer's method and chemical reduction of graphite oxide to graphene according to literature 35 . To prepare the graphene wrapped LiNi 0.5 Mn 1.5 O 4 nanorods, 0.05 g graphene and 1 g LiNi 0.5 Mn 1.5 O 4 nanorods were first mixed in a mortar for 0.5 h by grinding, and then dispersed in 30 ml ethanol by ultrasonication. After that, the dispersion was vigorously stirred at 50 °C for 8 h to achieve a Scientific RepoRts | 5:11958 | DOi: 10.1038/srep11958 uniform dispersion of LiNi 0.5 Mn 1.5 O 4 nanorods in the graphene matrix. Finally, the obtained mixture was dried in an oven at 80 °C overnight to obtain LiNi 0.5 Mn 1.5 O 4 -graphene composite.

Materials Characterization.
Structural features of the as-prepared LiNi 0.5 Mn 1.5 O 4 nanorods and the LiNi 0.5 Mn 1.5 O 4 -graphene composite were characterized with X-ray diffraction (XRD), Raman spectroscopy, and Fourier transform infrared (FTIR) spectroscopy. The XRD patterns were taken by a Shimadzu XRD-6000 X-ray diffractormeter with Cu Kα radiation between 10 and 80°. Raman spectra of different samples were acquired using a Renishaw in Via Reflex Raman microprobe with a 532 nm wavelength incident laser. FTIR spectra of the samples were collected from 800 to 400 cm −1 using a Nicolet-670 FTIR spectrometer. To determine the graphene content in the composite, thermogravimetric analysis (TGA) was carried out in the air at a heating rate of 10 °C min −1 from 30 to 700 °C using a DTG-60H Shimadzu thermal analyzer. The morphology features of the as-prepared samples were characterized with field-emission scanning electron microscopy (FESEM, JSM-6700F 15 kV), transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM, JEOL 2010 200 kV).

Electrochemical Measurements.
To investigate the electrochemical properties, half cells using lithium foil as both counter and reference electrodes were assembled with Lab-made Swagelok cells in an Ar-filled glove box. To prepare the working electrodes, active cathode materials (LiNi 0.5 Mn 1.5 O 4 nanorods and LiNi 0.5 Mn 1.5 O 4 -graphene composite), acetylene black (Super-P) and polyvinylidene fluoride (PVDF) with a weight ratio of 8:1:1 were mixed with N-methyl-2-pyrrolidinone (NMP) to form a slurry. The obtained slurry was coated onto the Al foils and dried at 120 °C for 2 h to remove the solvent. The electrodes were then pressed and cut into small disks (10 mm in diameter). The small disks were further dried at 80 °C in a vacuum oven for 12 h before battery tests. 1 M LiPF 6 in ethylene carbonate and diethyl carbonate (EC/DEC, v/v = 1:1) solution was used as the electrolyte and Celgard 2400 membrane was used as the separator. The galvanostatic charge/discharge measurements were carried out on a LAND CT2001A electrochemical workstation with a voltage window between 3.0 and 4.9 V (vs. Li + / Li) at different current densities (1 C is 148 mAh g −1 for LiNi 0.5 Mn 1.5 O 4 ) at room temperature. Cyclic voltammogram (CV) and electrochemical impedance spectroscopy (EIS) measurements were performed using a CHI660D electrochemical workstation. CVs were measured between 3.5 and 5.0 V (vs. Li/Li + ) at a scan rate of 0.05 mV s −1 . EIS measurements were carried out in the frequency range between 100 kHz to 0.01 Hz with an AC amplitude of 5 mV at an open circuit potential.