In-Situ Analysis on BaTiO3-LiCoO2 Composite Cathode for Lithium Ion Batteries

Abstract

The relatively low diffusivity of Li ions in bulk compounds has been limiting the achievable power densities for the lithium ion batteries (LIBs). Decorating of oxides such as Al2O3 [1], onto the active materials effectively improve the high rate capability. The thin-film oxide coating shortens the Li-ion diffusion distance at the SEI and prevents the dissolution of active material into the liquid electrolyte under a large current density. This phenomenon indicates the oxide thin films play a role as an artificial solid electrolyte interface (SEI). In our previous studies[2,3], the active material, LiCoO2 (LC) powder, was decorated with ferroelectric barium titanate, BaTiO3 (BT), as an artificial SEI, via a simple sol-gel route. The BT-loaded composite cathode exhibited a significant improvement in rate capability and an exceptionally high charge–discharge rate. However, the contribution of the ferroelectric BT SEI to such significant improvement in the cell performance hasn’t been clarified yet. The in-situ analysis, namely, electrochemical impedance spectroscopy (EIS) and X-ray absorption fine structure (XAFS) analysis, were performed in this study. BT-LC composite cathodes were prepared using a simple sol-gel synthesis. Various amount of BT (0.5, 1, 5, and 15 mol%) were loaded into the active materials LC(Cell Seed; Nippon Chemical Industrial Co., Ltd., Tokyo, Japan).Prior to impedance measurement, the cells were cycled in a voltage window of 3.3–4.5 V. During impedance measurements, cells were charged and discharged under constant current–constant voltage (CC–CV) with a voltage window of 3.3–4.5 V. During charging and discharging, the impedance was measured 12 times at 30-min intervals under open-circuit conditions. The impedance measurement was repeated for three cycles while charging and discharging at 0.1 C. As for the time-resolved dispersive XAFS (DXAFS) measurement, the loaded amounts of the oxides, BT and Al2O3, were fixed as 1 mol% in the LC matrix. Prior to DXAFS measurement, the cells were cycled by same way as impedance measurement. During DXAFS measurements, specimens were charged and discharged under a constant current (CC) with a cut-off voltage of 3.3 and 4.5 V for two cycles. The applied current density was fixed at a high rate (10 C). The acquired EIS data indicated the charge transfer resistance (R ct) of the BT 1mol% loaded specimen was 10.2Ω, corresponding to less than 1/12 the value for the bare LC (R ct = 127Ω).[4] This significant reduction in R ct may, therefore, be interpreted as the promotion of Li diffusivity at the electrode–electrolyte interface, driven by polarization due to the ferroelectricity of the BT layer. Meanwhile, the capacity retention ratio dropped rapidly, accompanied by a slight increase in R ct for larger BT loads of 5 and 15 mol %, attributed to the thicker BT layer and the existence of the impurity phase, BaCO3. These results imply that the ferroelectric BT SEI affected the kinetics of Li ion diffusion at the cathode–electrolyte interface, giving rise to a significant enhancement in the high charge–discharge rate capability, as well as restricting cobalt dissolution into the electrolyte. The DXAFS results also indicated, the BT 1mol%-loaded sample displayed the largest magnitude in Co valence variation during the cell driving at a high rate, while the energy shift for the Al2O3-coated sample was slightly larger than that of bare LC.[5] The increase in the X-ray absorption energy during the charging was intensified at higher applied potential (approaching 4.5 V) for the BT-decorated composite. The strength of the polarization increased with the applied potential during charging. The larger density of Li ions deintercalating from the LC, leading to higher capability, was attributed to the enhanced polarization at the composite–electrolyte interface. The heavier Co oxidation during charging, exhibited by the ferroelectric BT-SEI compared with the paraelectric Al2O3-SEI, was interpreted as evidence of the strengthened polarization due to the larger permittivity of BT. [1] Y. S. Jung, A. S. Cavanagh, A. C. Dillon, M. D. Groner, S. M. George, and S.-H. Lee, J. Electrochem. Soc., 157, A75-A81 (2010). [2] T. Teranishi, Y. Yoshikawa, R. Sakuma, H. Hashimoto, H. Hayashi, A. Kishimoto and T. Fujii, Appl. Phys. Lett., 105, 143904-1-3 (2014). [3] T. Teranishi, Y. Yoshikawa, R. Sakuma, H. Okamura, H. Hashimoto, H. Hayashi, T. Fujii, A. Kishimoto, and Y. Takeda, ECS Electrochem. Lett., 4, A137-A140 (2015). [4] T. Teranishi, Y. Yoshikawa, R. Sakuma, H. Okamura, H. Hayashi, A. Kishimoto, and Y. Takeda, Jpn. J. Appl. Phys., 54, 10NB02-1-5 (2015). [5] T. Teranishi, Y. Yoshikawa, R. Miyahara, , H. Hayashi, A. Kishimoto, M. Katayama, and Yasuhiro Inada, J. Ceram. Soc. Jpn., in press (2016).