Olivine phosphates have long been investigated as cathode material in Li-ion batteries. In particular, lithium iron phosphate (LiFePO4) has attracted the attention of many researchers because of its high theoretical capacity of 170 mAh g-1, low cost and high electrochemical/thermal stabilities of the phosphate (PO43-) anion[1]. However, the large volume difference during two-phase reaction between Li-rich Li1-aFePO4 (LFP) and Li-poor LibFePO4 (FP) phases leads to lattice mismatch at the LFP/FP boundary which induces a low Li-ion diffusion coefficient of 10-14 cm2 s-1 [2]. Such 1D Li+ diffusivity in the olivine-LFP is much slower (by 5-8 orders of magnitude) as compared with that of LCO (10-9 cm2 s-1) and LMO (10-5 cm2 s-1) cathodes. As a result, this slow lithium ion diffusion and the relatively poor electronic conductivity (10-10 to 10-7 Ω-1 cm-1) of the olivine-LFP in the absence of a doping cation, limit the power capability of the LFP material. To overcome these limitations, we have proposed new concept of three-phase structure containing a core of crystalline LFP, a shell of amorphous LFP, and graphitic carbon derived from Ketjen Black (KB) synthesized using our original in-situ ultracentrifugation process (UC process) [3]. The peculiar core-shell structure of LFP nanoparticle within the graphitic carbon prepared by UC-process improves the electronic conductivity in the whole LFP/graphitic carbon composites, while the amorphous LFP phase at the particle surface (shell) can achieve excellent rate performance owing to its high Li-ion diffusion coefficient.
2. Experimental
A precursor solution was prepared by mixing KB and H3PO4 aq. in ultra-pure water. The precursor solution was treated by UC process after an addition of iron acetate and lithium acetate. After drying at 80°C in vacuum for 12 h, the precursor composite powder was obtained. The powder was lastly annealed at 700 ◦C under N2 flow for 5 min and the final product (LFP/ graphitic carbon composite powder) was obtained. The LFP/graphitic carbon composite electrode was electrochemically characterized using a 2032 coin half-cell with Li metal in 1M LiPF6/EC+DEC (vol. 1:1).
3. Results and Discussion
The electron microscopy observations show that LFP/graphitic carbon composite has a highly crystalline phase of LFP core of ca.12-15 nm diameter with a distance of d=0.32 nm corresponding to (110) plane of the olivine LFP. This crystalline LFP phase is entirely covered with amorphous LFP. The combination of TEM observation together with XRD, XPS and Mössbauer analysis, supports the hypothesis that amorphous LFP contains Fe3+ defects. The most outer layer (“shell”) is composed of random graphitic carbon fragments/sheets stacked onto each other, derived from KB. The thickness of the shell is about 5 nm. An interlayer distance of d=0.35 nm was measured which is a little larger than that of graphene. This LFP/graphitic carbon composite enabled a 100C rate (36 seconds) discharge with 60 mAh g-1 per composite corresponding to 70% of the capacity obtained at the slowest discharge rate (1C). In the crystalline and amorphous LFP phase, different reaction mechanisms were observed and characterized by electrochemical study with a cavity microelectrode. While the reaction mechanism in the crystalline LFP phase is controlled by Li+ diffusion, the amorphous LFP phase shows a fast, surface-controlled, pseudocapacitive charge-storage mechanism[4]. This pseudocapacitive behavior is extrinsic in origin since it comes from the presence of Fe3+ defects in the structure. These features explain the ultrafast performance of the material which offers interesting opportunities as a positive electrode for assembling high power and high energy hybrid supercapacitors.
References
1) A. K. Padhi, et al., J. Electrochem. Soc., 1997, 144, 1188-1194.
2) G. Kobayashi, et al., Adv. Funct. Mater., 2009, 19, 395-403.
3) K. Naoi, et al., Energy Environ. Sci., 2016, 9, 2143-2151.
4) K. Kisu, et al., Electrochem. Commun., 2016, 72, 10-14.