Li1.2Mn0.54Ni0.13Co0.13O2-Encapsulated Carbon Nanofiber Network Cathodes with Improved Stability and Rate Capability for Li-ion Batteries

Li1.2Mn0.54Ni0.13Co0.13O2-encapsulated carbon nanofiber network cathode materials were synthesized by a facile electrospinning method. The microstructures, morphologies and electrochemical properties are characterized by X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM), high resolution transmission electron microscopy (HR-TEM), galvonostatic charge/discharge tests, cyclic voltammetry and electrochemical impedance spectroscopy (EIS), etc. The nanofiber decorated Li1.2Mn0.54Ni0.13Co0.13O2 electrode demonstrated higher coulombic efficiency of 83.5%, and discharge capacity of 263.7 mAh g−1 at 1 C as well as higher stability compared to the pristine particle counterpart. The superior electrochemical performance results from the novel network structure which provides fast transport channels for electrons and lithium ions and the outer carbon acts a protection layer which prevents the inner oxides from reacting with HF in the electrolyte during charge-discharge cycling.

capability and more stable cyclic performance compared to the pristine counterpart. The results not only give a further insight into the reaction mechanism but also provide a rational direction to synthesize cathode materials for Li-ion batteries. Figure 1 shows the XRD patterns of the pristine particle and nanofiber decorated samples. The pristine particle sample shows the α -NaFeO 2 structure without any impurity reflections. The weak superstructure reflections around 2θ = 20°-25° belongs to the ordering of the transition metal and Li ions in the transition metal layer of the lattice 25,26 . Clear separation of adjacent peaks of (006)/(012) and (108)/(110) indicates that the samples have a well crystalline layered structure [27][28][29] . For the nanofiber decorated sample, no significant lattice parameter differences were recognized in the main reflections corresponding to the α -NaFeO 2 structure, and no additional reflections corresponding to carbon are observed due to its amorphous structure 21,22 .

Results and Discussion
The morphologies and strucutres of the nanofiber decorated sample were shown in Fig. 2. Compared to the pristine particles (Supporting Information, Figure 1S), which showed a universal and slight agglomeration with a uniform particle size ranging between 300 and 500 nm, the carbon nanofiber contained Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 particles which were evenly dispersed with carbon black, excepting a little agglomeration after electrospinning (Fig. 2a). The morphology of nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 after heat treatment in air was shown in Fig. 2b. The nanofibers shrinked and partly broken, some particles were inevitably squeezed out from the nanofibers and then distributed along the surface or junction of the nanofibers after agglomeration. Figure 2c,d show the HRTEM images of the samples. Result (Fig. 2c) confirms that Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 particles and carbon black are dispersed inside the carbon nanofiber matrix. The particles exhibit continuous interference fringe spacing of around 0.47 nm corresponding to the (003) lattice fringes of the hexagonal layered structure (Fig. 2d) 3,6,28,29 . In addition, amorphous carbon layers with a thickness about 10 nm were observed on the surface of the Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 particles, which may be regarded as protection layer during electrochemical cycling. The selected area electron diffraction (SAED) pattern (inset of Fig. 2d) confirms the single crystallinity for particles.
X-ray photoelectron spectroscopy (XPS) measurements were used to investigate the variation in chemical states of Mn, Ni and Co elements. Figure 3 shows typical XPS spectra for Mn2p, Co2p and Ni2p for the pristine and nanofiber decorated samples. For these two samples, the observed binding energies of Mn2p, Co2p and Ni2p remain at 642, 780 and 855 eV, which coincide well with those for Mn 4+ , Co 3+ and Ni 2+ , respectively 30,31 . Therefore, the valences of Mn, Co and Ni did not change after the heat treatment with/without electrospinning.
The first charge-discharge curves during electrochemical cycling of the two cathode materials taken at 0.2 C, are shown in Fig. 4a. During the charging process, both samples exhibit two distinguishable stages, a smoothly sloping voltage profile below 4.5 V corresponding to Li + deintercalation from LiMn 1/3 Ni 1/3 Co 1/3 O 2 component and a long plateau around 4.5 V related to Li + and O 2− extracted from the Li 2 MnO 3 phase and structural rearrangement 1,31-33 . The first discharge capacity of nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 /C and pristine sample are 263.7 and 247.2 mAh g −1 , with the coulombic efficiency of 83.5% and 75.4%, respectively. The higher coulombic efficiency and discharge capacity may be attributed to the reasons as follow: (a) the amorphous carbon layer prevents the formation of inactive O 2 molecules and the side reactions of electrolyte; (b) the better dispersity and less agglomeration of the    Figure 4b shows the CV profiles of the first cycle of the two samples at a sweep rate of 1 mV s −1 between 2.0-4.8 V. As can be seen, the initial anodic peaks of the samples at about 4.2 V are due to the lithium extraction from the Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 lattice and the oxidation process of Ni 2+ /Co 3+ to higher oxidation states 2,30,34-36 , while the sole cathodic peak occurs at around 3.8 V is attributed to the reduction of Ni 2+ /Co 3+ . The second anodic peaks located in the potential of 4.6 V are normally associated with the oxygen release process from Li 2 MnO 3 -type component as well as the electrolyte oxidation. Compared with the CV curve of pristine Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 , the decreased intensity and area of anodic peak around 4.6 V of the nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 sample implies that the O 2 release was greatly suppressed by the amorphous carbon layers and more available Li + sites were remained. This agrees with the improved columbic efficiency of the composite sample (Fig. 4a).
The cyclic performances of both samples at 1 C are shown in Fig. 5. The first discharge capacity of nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 and particle samples are 219.5 and 205 mAh g −1 with a coulombic efficiency of 82.5% and 76.2%, respectively. After 100 cycles, the nanofiber samples still display a  discharge capacity of 176.7 mAh g −1 with a capacity retention of 80.5%, while the particle samples only show a discharge capacity of 145.9 mAh g −1 and a capacity retention of 71.1%. Their cyclic stability at low rate of 0.1 and 0.2 C are also shown in the inset of Fig. 5. The two samples deliver a similar discharge capacity in the first several cycles. However, the nanofiber samples still retain a higher discharge capacity than the pristine one both at 0.1 and 0.2 C after 100 cyles. The improved cyclic stability and capacity retention of the nanofiber cathodes may be attributed to the synergistic effect of the larger specific surface area and the protection of coating layer. Comparing with the pristine samples, the larger specific surface area of nanofibers provides more transportation paths for ions and/or electrons, and the coating layer simultaneously prevents the inner oxide from reacting with HF in the electrolyte and twice agglomeration during electrochemical test. Figure 6 shows a continuous cycling result at incremental rates from 0.2 to 5 C then recovering back to 0.2 C. Similar discharge capacity is observed at lower rate for both samples, but obvious capacity differences increase gradually with the increasing rate. As can be seen, the rate capability for nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 samples has been significantly enhanced compared to the particle. Such contrast observation might be related to the faster ionic and/or electronic diffusion and shorter migration path at high rate, which were benefited from the network structures.
To investigate the origin of the improved electrochemical performance of nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 sample, EIS plots of both samples were collected. Figure 7 compares the EIS spectra of the particle and nanofiber samples at different states of cycling, 1st and 10th. An equivalent circuit (inset of Fig. 7) 28 was applied to fit the raw data from EIS measurements to obtain the accurate values of different resistances. In this equivalent circuit, R Ω is the resistance of the solution , R s is the resistance of ion diffusion in the region of the surface layer of particle, R ct represents the charge transfer resistance in the electrode/electrolyte interface and Z w refers to Warburg impedance .The fitting values of R s and R ct were summarized in Table 1.
Compared with the Fig. 7a,b, there was a huge reduction of R ct for the nanofiber sample upon cycling. After 10 cycles, the R ct values of nanofiber and particle samples increased from 21.73 and 30.91 Ω to 118.26 and 363.31 Ω , respectively. The reduction of R ct for nanofiber sample may facilitate lithium transfer on the electrode/electrolyte interface, then leads to improved rate capability. Therefore, the better electrochemical performance of nanofiber sample compared to the particle one can be ascribed to larger surface area and better electrical contact of the particle, which increase the active specific surface area and decrease the polarization effect.

Conclusions
In conclusion, Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 -ecapulated carbon nanofiber network were prepared through a facile electrospinning method following by a heat treatment. XPS spectra demonstrate that the valence of Mn, Ni, Co did not change after the heat treatment. Compared with the pristine Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 particles, the nanofiber sample has the obviously improved electrochemical performance. The coulombic efficiency of the first charge-discharge process is enhanced from 76.2% to 82.5% with a capacity retention of 80.5% at 1 C after 100 cycles. In addition, the nanofiber sample also exhibits superior rate capability compared to the pristine one. Such improved initial coulombic efficiency, cyclic stability and rate capability can be attributed to the suppression of the side reaction of electrolyte, larger surface area of nanofibers, better dispersity of active particles and more importantly, faster electron and ion transportation, which are benefited from the carbon nanofiber network. Therefore, it is believed that the nanofiber decorated Li 1.2 Mn 0.54 Ni 0.13 Co 0.13 O 2 should be one of ideal cathode candidates and the results also provide a rational direction to synthesize other cathode materials for Li-ion batteries. Then the solution was evaporated at 85 °C until a viscous violet gel emerged. The gel was dried in a vacuum oven at 120 °C for 18 h. The precursor powders were decomposed at 500 °C for 6 h to guarantee the complete decomposition of (PVP). After a grinding for 15 min with a mortar and pestle, the obtained powders were calcined at 950 °C for 8 h in air and ground again for at least 20 min to get the finial products.     . Both of the Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) (at a frequency range of 10 −2 -10 5 Hz) were conducted using an electrochemical workstation (CHI660A), and the CV was applied with a scan rate of 1 mV/s between 2.0-4.8 V. All of the electrochemical tests were carried out at room temperature.