摘要
Lithium-ion batteries (LIBs) are one of the most promising energy storing devices, offering high volumetric and gravimetric energy density compared with other battery technologies [1]. However, LIBs are suffering from safety issues due to the poor chemical stability and the flammability of organic liquid electrolytes. Solid state electrolytes are expected to solve such safety issues. Since the initial study by Murugan et al. in 2007, Li 7 La 3 Zr 2 O 12 (LLZO) garnet has received much attention due to its high ionic conductivity (10 -3 to 10 -4 S cm -1 ) and stability against Li metal. [2-3]. LLZO is crystallized in both cubic (space group Ia-3d) and tetragonal (space group I4 1 /acd) form. The ionic conductivity of cubic phase is two order magnitudes higher than that of tetragonal phase [3]. However, it is difficult to obtain the cubic phase LLZO because the cubic phase is not stable at room temperature and can be obtained at high sintering temperature (>1200 °C). Therefore, super-valent cation doping, such as Al 3+ and Ga 3+ , is introduced in order to increase the number of vacancies and results in the enhancement of stability of the high conductivity cubic phase at room temperature. Ga 3+ is one of the most promising candidates due to its high ionic conductivity. Both grain size and density are important factors in determining the ionic conductivity, but the relationship is not clear in Ga-doped LLZO. In this study, we investigated the correlation between grain size and ionic conductivity in order to understand how to improve the ionic conductivity of Ga-doped LLZO. The grain size of Ga-doped LLZO has been controlled by changing the sintering temperature (1000 °C and 1150 °C for 8 h) and employing the two-step sintering method (1150 °C for 5 min-1000 °C for 8 h). We synthesized the Ga 3+ -doped LLZO (Li 6.1 Ga 0.3 La 3 Zr 2 O 12 ) from LiOHH 2 O, La 2 O 3 , ZrO 2 , and Ga 2 O 3 . The 15 wt% excess of LiOHH 2 O was used to compensate the lithium loss during high temperature sintering. The powders were ball-milled in isopropyl alcohol for 12 h at a milling speed of 250 rpm/min and dried for 12 h. The well-ground powder was calcined at 600 °C for 6 h in alumina crucible and then reground, pelletized, and sintered at temperatures for 8 h covered with mother powders to prevent Li loss during sintering. Li + evaporation typically leads to the pyrochlore La 2 Zr 2 O 7 phase formation. The relative density of the pellets was obtained as 93.2, 93.4, and 94.5% for 1000 °C sintered LLZO (Ga-LLZO-1000), 1150 °C sintered LLZO (Ga-LLZO-1150), and two-step sintered LLZO (Ga-LLZO-TS), respectively. The total ionic conductivity of Ga-LLZO-1150 was much higher (1.22 x 10 -3 S cm -1 ) than that of Ga-LLZO-1000 (3.05 x 10 -4 S cm -1 ) although the relative density of both samples was similar. The grain coarsening was observed in Ga-LLZO-1150 with size of >200 µm whereas the grain size of Ga-LLZO-1000 was <10 µm (Fig. 1a and 1c). Ga-LLZO-TS had the highest relative density with a medium grain size (100-200 µm), but showed the lower total ionic conductivity (9.91 x 10 -4 S cm -1 ) than Ga-LLZO-1150 (Fig. 1b). In general, the increase of density leads to the higher ionic conductivity of materials because higher density provides more pathways for lithium ions transport across the material. For comparison, the ionic conductivity and the density of undoped and Al-doped LLZO were also measured, and the ionic conductivity of both samples increased with increasing the density or sintering temperature. However, this was not true in Ga-doped LLZO. The conductivity of the specimen with higher density and smaller grain size was lower than that of lower density with larger grain size. Both bulk and grain boundary contribute to the total ionic conductivity. The total ionic conductivity (σ t otal ) is calculated according to σ t otal = d/(A x R total ), where d is the pellet thickness, A is the area of the electrode and R total represents the total resistance of the pellet which is the sum of R bulk and R grain boundary . Ga-LLZO-1150 with large grain size has the higher ionic conductivity because the larger grain size leads to an enhancement in the ratio of grain/grain boundaries, resulting the improvement of the total ionic conductivity by reducing the grain boundary resistance. The ionic conductivity of Ga-doped LLZO was more affected by the grain size than the density. Reference [1] J.-M. Tarascon, M. Armand, Nature 414 (2001) 359–367. [2] R. Murugan et al., Angew. Chem. (2007) 119, 7925. [3] H. Buschmann et al., Phys. Chem. Chem. Phys. 13 (2011) 19378–19392. Figure 1