Stable LATP/LAGP double-layer solid electrolyte prepared via a simple dry-pressing method for solid state lithium ion batteries

Abstract

In this paper, a NASICON-type Li_1.3Al_0.3Ti_1.7(PO₄)₃ (LATP)/Li_1.3Al_0.3Ge_1.7(PO₄)₃ (LAGP) bi-layer structured solid state electrolyte was successfully prepared via a simple dry pressing and post-calcination method. By adjusting the sintering temperature for LAGP starting materials, a dense and smooth LATP/LAGP double-layer solid state electrolyte with no defects was obtained. This electrolyte sample exhibits a high electrical conductivity of 3.4 × 10⁻⁴ S cm⁻¹ and a negligible electronic conductivity of 9.6 × 10⁻⁹ S cm⁻¹ at room temperature. In addition, the LATP/LAGP electrolyte also shows an excellent stability in air as well as chemical stability against Li. Moreover, an assembled LiFePO₄/LATP–LAGP/Li coin-type battery employing LATP/LAGP as the solid state electrolyte can be suitably charged and discharged at a current rate of 0.1C at room temperature, and its low charge–discharge capacities are mainly attributed to the high electrolyte/electrode interfacial resistance of the cell. These results suggest that the LATP/LAGP bi-layer electrolyte can be an alternative electrolyte for all-solid-state lithium-ion batteries.

---

Introduction

Recently, all-solid-state rechargeable lithium ion batteries (LIBs) with solid electrolytes have attracted a lot of attention due to their high energy density, good safety as well as long cycle performance, and can be applied to some large-scale systems such as electric or hybrid vehicles and smart grids. The key to the fabrication of high-performance all solid state LIBs is to develop an appropriate solid state electrolyte, which should possess the following properties: high lithium ion conductivity, a wide electrochemical window, excellent chemical stability with lithium anodes and Co-, Ni-, or Mn-containing cathodes. So far, various types of lithium ion conducting solid state electrolytes have been developed and investigated as possible electrolyte candidates for all solid state LIBS, which mainly include garnet-type Li₇La₃Zr₂O₁₂,^1,2 Li₂S-based glass,^3–5 NASICON-type LiM₂(PO₄)₃ (M = Ti, Ge, Hf, Zr, Sn).^6–8

Among the above-mentioned solid state electrolytes, Li₇La₃Zr₂O₁₂ and Li₂S-based glass have been widely used to fabricate all solid state LIBS. However, these two electrolytes still have some drawbacks. For Li₇La₃Zr₂O₁₂ electrolyte, the high calcination temperature up to 1200 °C is required to obtain the cubic phase with high lithium ion conductivity. Under such circumstance, the La₂Zr₂O₇ impurity is easily formed, and thus reduces lithium ion conductivity. In addition, when Li₇La₃Zr₂O₁₂ sample is exposed to the humid air, its grain boundary conductivity obviously decreased due to the formation of La(OH)₃ phase in Li₇La₃Zr₂O₁₂ grain boundaries.⁹ As to Li₂S-based glass, it is very instable in air, which tends to react with the water contained in the air, leading to the formation of harmful H₂S gases.^10,11 Compared with the above two electrolytes, the NASICON-type LATP solid state electrolyte has some advantages over them, such as excellent stability in ambient atmosphere, mass production and low cost. However, the LATP electrolyte tends to react with lithium anode due to the Ti⁴⁺/Ti³⁺ redox reaction.^12,13

The problem for LATP solid electrolyte can be solved by adopting the strategy that a barrier layer, which requires sufficient stability against lithium metal, can be introduced between electrolyte and lithium anode. The added barrier layer can block the contacts between them and prevent the undesired chemical reactions. This method has been employed to prepare all solid state LIBS with the aim of avoiding the diffusion of elements at the electrode/electrolyte interface or lowering the electrode/electrolyte interfacial resistance.^14–16 In the previous work, the interlayers were prepared using radio-frequency (RF) reactive magnetron sputtering. Although this technique can be used to fabricate dense and high performing interlayers, it is expensive, complex and consumes a large amount of energy. Therefore, it is not suitable for the mass production.

To significantly lower the cost of production, the dry pressing method can be adopted to produce the interlayer for all solid state LIBS, which is simple, reproducible, and very cost-effective as compared to vacuum deposition techniques, and it has been widely applied to the fabrication of electrolyte membranes and interlayers at the interface of electrolyte/electrode for solid oxide fuel cells.^17–20 To the best of our knowledge, there were no reports on the preparation of protecting layer between LATP electrolyte and lithium anode using the dry pressing method. In this work, the NASICON-structured LAGP solid electrolyte was chosen as the materials for the preparation of interlayer, whose lithium ion conductivity can reach over 10⁻⁴ S cm⁻¹ and is slightly lower than that of LATP sample.^21,22 Importantly, comparable with LATP electrolyte, the LAGP solid state electrolyte is more stable against Li because Ge⁴⁺ is more difficult to be reduced than Ti⁴⁺.^23–25 Taking into account of the high price for Ge sources, it is not practical to use LAGP directly as solid electrolyte for all solid state LIBS, while as a interlayer, only a very small amount of LAGP materials is needed, which can reduce the low preparation cost of all solid state LIBS. Herein, a simple dry pressing and post-calcination process was employed to fabricate LATP/LAGP double-layer lithium ion solid state electrolyte. The electrical conduction properties, stability in air and chemical stability against Li of LATP/LAGP electrolyte as well as electrochemical characteristics of lithium ion battery employing LATP/LAGP as solid state electrolyte were investigated.

Experimental section

Synthesis of LATP and LAGP electrolyte materials

The LATP electrolyte materials were prepared via a simple solution method.²⁶ The chemicals used in this synthesis were LiOH·H₂O (99.99%, Aladdin), Al₂O₃ (99%, Sinopharm), TiO₂ (99%, Aladdin), and H₃PO₄ (85%, Sinopharm). The typical synthesis process was as follows. LiOH·H₂O was firstly dissolved in deionized water, and then Al₂O₃, TiO₂, and H₃PO₄ were added to the resultant solution, respectively. In order to compensate volatile Li components, a 10 wt% excess of LiOH·H₂O was used during the preparation process. The obtained precursor solution was dried at 180 °C in an oil bath to remove the excessive water until a viscous paste was formed. Then, the above paste was put in an aluminum crucible and calcined in a box furnace at 700 °C for 4 h at a heating rate of 2 °C min⁻¹ to ensure the formation of the crystalline LATP powders. The as-synthesized powders were further ball-milled in an ethanol medium using a planetary mill at 450 rmp for 10 h in order to get the fine powders. Based on the similar synthesis process, the LAGP powders were also prepared.

Preparation of LATP/LAGP double-layer electrolyte

Subsequently, the LATP/LAGP double-layer solid state electrolyte was prepared by a dry-pressing and post-calcination method. First, the LATP powders were uniaxially pressed under 8 MPa pressure as a substrate in a stainless-steel die. Then, a very small amount of LAGP powders was uniformly distributed onto the LATP substrate and co-pressed at the pressure of 20 MPa for 5 min to form a LATP/LAGP double-layer green pellet. The bi-layer pellet was subsequently sintered at 750 °C for 5 h, and thus a dense, well-bonded LATP/LAGP electrolyte was obtained. The synthetic process was schematically illustrated in Fig. 1.

Fig. 1 Schematic illustration of the preparation process for LATP/LAGP bi-layer electrolyte.

Characterization of LATP/LAGP double-layer electrolyte

The surface and cross-sectional microstructure of LATP/LAGP double-layer electrolyte was observed by a scanning electron microscope (SEM, HITACHI S-4800). Elemental distribution was analyzed using Energy Dispersive X-ray detector equipped on SEM. The phase composition of the as-obtained double-layer electrolyte was examined by X-ray diffraction (XRD, Bruker-AXS Microdiffractometer D8 Advance, Cu Ka) in the 2θ range from 5° to 90° with a step size of 0.02° at ambient temperature. The electrical conductivity was measured using AC impedance technique with Li-ion blocking Au electrodes which were sputtered onto both sides of LATP/LAGP electrolyte. This measurement was performed using an electrochemical work station (CHI 660E) in the temperature range of 25–105 °C with the frequency ranging from 1 Hz to 100 kHz. The electronic conductivity was estimated using a potentiostatic polarization method under a polarizing voltage of 0.2 V. The stability in air for LATP/LAGP electrolyte was evaluated by observing the evolution of impedance spectra as function of dwell time in air. To evaluate the chemical stability between LATP–LAGP and Li anode, a LiFePO₄/LATP–LAGP/Li all-solid-state cell was assembled, and the variation of its electrochemical impedance spectroscopy with attachment time was investigated. The typical preparation process of LiFePO₄ composite cathode can be found in reference.²⁷

Preparation and property of LiFePO₄/LATP–LAGP/cell

A LiFePO₄/LATP–LAGP/Li coin cell was also constructed in order to verify whether the LATP/LAGP double-layer solid electrolyte can be applied to all solid state LIBS. The typical preparation process of cell was follows. LiFePO₄ active materials, poly-vinylidene fluoride (PVDF) and super P conductive additive with the mass ratio of 8 : 1 : 1 were uniformly mixed and subsequently ground with the appropriate amount of N-methyl-2-pyrrolidon (NMP) to form the cathode paste. Then, the slurry was cast on an aluminum foil and vacuum dried at 120 °C for 12 h to obtain the cathode. The lithium foil was used as anode. The LATP/LAGP pellet was used as electrolyte. To reduce the cathode/electrolyte interfacial resistance, a small amount of liquid electrolyte (1 M lithium hexafluorophosphate (LiPF₆) solution dissolved in ethylene carbonate (EC) and dimethyl carbonate (DMC) (EC : DMC = 1 : 1)) was added between them. A 2032-type coin cell was assembled in an argon-filled glove box. Its charge/discharge performance was evaluated on a LAND-2010 automatic battery tester between 2.5 and 4.0 V at a constant current density of 0.1C.

Results and discussion

The effect of calcination temperature for LAGP materials on microstructure of LATP/LAGP bi-layer electrolyte obtained via a dry-pressing and post calcination method was examined by SEM. Fig. 2 shows the surface SEM images of LATP/LAGP bi-layer structures made from LAGP powders calcined at different temperatures.

Fig. 2 The surface SEM images of LATP/LAGP bi-layer structures made from LAGP powders calcined at different temperature: (a) 650 °C; (b) 700 °C; (c) 800 °C.

As shown in Fig. 2(a), a smooth and flat LAGP surface with no defects was obtained when the LAGP powders calcined at 650 °C were used. As the calcination temperature of LAGP starting powders increases, the cracks were observed on the LAGP electrolyte surfaces and the number of defects significantly increased, which can be seen from Fig. 2(b) and (c). The sintering mismatch between the LAGP layer and the underneath LATP substrate can cause a large tensile stress on the surface of LAGP layer, resulting in the formation of defects. Therefore, the optimum calcination temperature for LAGP materials was fixed at 650 °C.

Fig. 3 presents the cross-sectional SEM micrographs of the LATP/LAGP bi-layer electrolyte and the corresponding elemental distribution maps of O, P, Al, Ti and Ge. There is no delamination between the LAGP layer and the LATP substrate. The LAGP layer has a thickness of 55 μm, which is tightly adhered to the LATP substrate, as displayed in Fig. 3(a). Moreover, the LAGP layer was sintered well so that no obvious defects such as voids and cracks were found (shown in Fig. 3(g)). Also, the EDS mappings show that O, P and Al are homogenously distributed over the entire fracture surface. Ge was also confirmed and only found in the LAGP layer, while Ti is just located in the LATP substrate, implying that there is no elemental diffusion between these two layers.

Fig. 3 The cross-sectional SEM micrographs of the LATP/LAGP bi-layer electrolyte (a and g) and the corresponding elemental distribution maps of O, P, Al, Ti and Ge (b–f).

To confirm the phase compositions of LATP/LAGP electrolyte, the phases of LAGP and LATP layers were characterized by XRD, respectively. Fig. 4 shows the XRD patterns of LAGP and LATP layers calcined at 750 °C for 5 h, in which all diffraction peaks for LATP layer were well-matched with the standard pattern of LiTi₂PO₄ (JCPDS 35-0754) with the space group of R3c. For the LAGP layer, its diffraction peaks corresponded to the LAGP crystalline phase except some small peaks attributed to impurities of Li₂O.

Fig. 4 The XRD patterns of LAGP layer and LATP substrate calcined at 750 °C for 5 h.

Based on Au/LATP–LAGP/Au pellet, the electrical conductivity of LAGP/LATP bi-layer electrolyte was evaluated using AC impedance technique, and the result was displayed in Fig. 5.

Fig. 5 The typical Nyquist plot of LAGP/LATP electrolyte measured at 25 °C.

As shown in Fig. 5 is a typical Nyquist plot of LAGP/LATP electrolyte measured at 25 °C, in which a high-frequency semicircle and a low-frequency inclined line are observed. The semicircle represents the grain boundary impedance, and the straight line corresponds to Warburg-type impedance which originates from the diffusion of lithium ions in the gold blocking electrodes. The first intercept of the semicircle on the real axis at the high frequency region represents the bulk resistance and the second one in the low frequency region corresponds to the total resistance.²⁸ To obtain the value for the total resistance, the impedance spectra was fitted using an equivalent circuit model, R_g(CPE₁R_gb)CPE₂ where R_g is a grain resistor, R_gb is a grain boundary resistor, and CPE is a constant phase element. And the fitted values for bulk resistance, grain boundary resistance and total resistance are 47.3 Ω, 109.2 Ω and 156.5 Ω, respectively. Subsequently, the electrical conductivity of LATP/LAGP electrolyte was calculated from the following equation: σ_t = (1/R_t)(l/S), where σ_t, l, R_t and S denote the electrical conductivity, thickness, the total resistance, and effective area of LATP/LAGP electrolyte, respectively. The electrical conductivity of LATP/LAGP electrolyte measured at 25 °C was calculated as 3.4 × 10⁻⁴ S cm⁻¹, which is comparable to those of LATP or LAGP electrolytes reported in the literatures,^{21,22,29–33} indicating that the LATP/LAGP bi-layer electrolyte can meet the requirement of all solid state LIBS for electrical conductivity. For comparison, the LATP pellets and LAGP pellets were also prepared. The densities of LATP pellets are 2.41 g cm⁻³ and 2.74 g cm⁻³ before and after calcination, which correspond to 84.2% and 94.0% of the theoretical density. For LAGP pellets, their densities are 2.84 g cm⁻³ and 3.18 g cm⁻³, which are 83.5% and 93.5% of the theoretical density. Subsequently, the impedance spectra of LATP and LAGP electrolytes were also measured at 25 °C and shown in Fig. S1,† and the corresponding electrical conductivities are 7.0 × 10⁻⁴ S cm⁻¹ and 1.6 × 10⁻⁴ S cm⁻¹, respectively. Therefore, the electrical conductivity of LATP/LAGP electrolyte was lower than LATP component but higher than LAGP component.

The complex impedance spectra of LATP/LAGP sample measured in the range of 303 K to 378 K are shown in Fig. 6(a). For LATP and LAGP samples, their impedance spectra are displayed in Fig. S2(a) and (b).† The semicircles in complex impedance spectra are reduced with increase of measurement temperatures, which can be attributable to faster lithium ion conduction at higher temperature. The Arrhenius plots for the relationship between the total electrical conductivity and the inverse temperature are given in Fig. 6(b) and S2(c).† The activation energy for the total conductivity was estimated according to the Arrhenius equation σ_TT = A exp(−E_a/kT), where E_a, A, k, and T denote the activation energy, the pre-exponential factor, the Boltzmann's constant and the absolute temperature, respectively. The values of activation energy for LATP/LAGP, LATP, and LAGP samples were determined to be 0.32 eV, 0.27 eV, and 0.33 eV, respectively, which suggests that there exists high ion mobility in the lattice and the grain boundaries.

Fig. 6 (a) The complex impedance spectra of LATP/LAGP samples measured in the range of 303 K to 378 K; (b) the Arrhenius plots for the relationship between the total electrical conductivity and the inverse temperature.

To check the contribution of electronic conductivity to total conductivity, a potential static polarization method was carried out to measure the electronic conductivity of LATP/LAGP bi-layer electrolyte. The current dependence of polarization time is displayed in Fig. 7. As can be found, as the polarization time increases, the current reaches the steady state. Based on the steady state current, the electronic conductivity can be calculated from the following equation: σ_e = 4lI/(πD²U) where U is the polarization voltage, l is the thickness of the electrolyte, D is the diameter of Au electrode, and I is the current, respectively. The electronic conductivity of LATP/LAGP bi-layer electrolyte is 9.6 × 10⁻⁹ S cm⁻¹, which is at least 5 orders of magnitude lower than the total conductivity, suggesting that the electronic conduction contribution can be negligible.

Fig. 7 The evolution of current with polarization time under a polarization voltage of 0.2 V.

Due to the preparation of starting powders and LATP/LAGP bi-layer electrolyte in air atmosphere, it is necessary to evaluate the stability of LATP/LAGP bi-layer electrolyte in air.

Fig. 8 shows the impedance spectrum of LATP/LAGP bi-layer electrolyte measured before and after 3 months of exposure to air. No obvious changes in impedance plots were found after 3 months, which demonstrates that the LATP/LAGP electrolyte has an excellent stability in ambient atmosphere.

Fig. 8 The impedance spectrum for LATP/LAGP bi-layer electrolyte measured before and after 3 months of exposure to air.

For a solid state electrolyte adoptable in all solid state LIBS, it requires sufficient chemical stability against lithium metal. Therefore, the chemical stability between LATP/LAGP bi-layer electrolyte and lithium metal was investigated by comparing the evolution of impedance spectra for LFP/LATP–LAGP/Li structured cell. Herein, the substitution of Li metal by LFP is to avoid the undesired reaction between Li metal and LATP layer. As shown in Fig. 9, the impedance spectrum of the cell almost kept unchanged after four weeks, indicating that no obvious reaction at the LATP–LAGP/Li interface takes place. This result implies that the LATP–LAGP sample is stable against Li mental.

Fig. 9 The evolution of impedance spectra for the LFP/LATP–LAGP/Li structured cell.

To further examine whether the electrolyte sample can be applied to all solid state LIBS or not, a coin-type cell with LFP/LATP–LAGP/Li structure was assembled and its charge–discharge performance was evaluated at room temperature. To reduce the LFP/LATP–LAGP interfacial resistance, a small amount of liquid electrolyte was introduced at the LFP/LATP–LAGP interface. Fig. 10 shows the galvanostatic charge–discharge profiles of LFP/LATP–LAGP/Li cell within the voltage range of 2.5–4.0 V at the current rate of 0.1C. A voltage plateau was observed at about 3.55 V, which is higher than the typical extraction/insertion reaction of LFP (3.45 V). This phenomenon can be attributed to the large interfacial polarization. The initial charge and discharge capacities of this battery are 142.0 mA h g⁻¹ and 130.6 mA h g⁻¹, respectively, corresponding to approximately 83.5% and 76.8% of theoretical capacity (170 mA h g⁻¹). After 3 charge–discharge cycles, the charge and discharge capacities increased to 145.3 and 141.4 mA h g⁻¹, respectively. The low capacity of cell was ascribed to the high electrolyte/electrode interfacial resistance of the cell, which can be confirmed from the impedance spectra of the cell shown in Fig. 11. Based on the equivalent circuit of R₀(CPE₁R₁)(CPE₂R₂)W in the inset of Fig. 11, the electrochemical impedance spectra was analyzed. The LFP/LATP–LAGP/Li cell exhibits two semicircles and an inclined line. The first semicircle at high-frequency region corresponds to the grain boundary resistance (R₁) of LATP/LAGP electrolyte, while the second one at middle-frequency can be assigned to the electrolyte/electrode interfacial resistance (R₂). The first intercept of the first semicircle on the real axis reflects the bulk resistance (R₀) of LATP/LAGP electrolyte. The fitted values of R₀, R₁ and R₂ are 36.4 Ω, 128.3 Ω and 4121 Ω, respectively, and the value of electrolyte resistance (R₀ + R₁) of the cell is 164.7 Ω, which is almost consistent with that of LATP/LAGP electrolyte obtained from Fig. 5. Therefore, it is concluded that the electrolyte/electrode interfacial resistance, which originates from the poor contacts at the electrolyte/electrode interfaces, plays a dominant role in the internal resistance of the cell. These results primarily demonstrates that the LATP/LAGP bi-layer structure can be used as a solid state electrolyte for all solid state LIBS. In the future work, we will improve the performance of the cell by optimizing the electrolyte/electrode interfaces.

Fig. 10 The galvanostatic charge and discharge curves of the LFP/LATP–LAGP/Li structured cell at 0.1C at room temperature.

Fig. 11 The impedance spectra of the LFP/LATP–LAGP/Li structured cell measured at 25 °C.

Conclusions

In summary, a facile dry pressing and subsequent calcination process has been adopted to prepare the dense, smooth and flat LATP/LAGP bi-layer solid state electrolyte which shows a high electrical conductivity of 3.4 × 10⁻⁴ S cm⁻¹ at 25 °C as well as a negligible electronic conductivity. This electrolyte was also found to be excellent stability in air, with no performance decay after exposure to air atmosphere for 3 months. Simultaneously, the LATP/LAGP bi-layer electrolyte is stable in contact with Li mental, which is attributed to good stability of LAGP layer against Li anode. The LFP cell assembled by LATP/LAGP electrolyte can be normally charged and discharged at 0.1C rate. After 3 charge–discharge cycles, the cell capacity increases to 145.3 mA h g⁻¹ and 141.1 mA h g⁻¹, corresponding to about 85% and 83% of theoretical capacity. These results confirms that the LATP/LAGP bi-layer electrolyte can be applied to all solid state LIBS. In our future work, we will assemble all solid state LIBS using the above electrolyte and evaluate their electrochemical performances.

Acknowledgements

The authors acknowledge the financial support of the “100 Talents” program of Chinese Academy of Sciences (Y51002110W) and the Qingdao Key Laboratory of Solar Energy Utilization and Energy Storage Technology.

Notes and references

1. M. Huang, T. Liu, Y.-F. Deng, H.-X. Geng, Y. Shen, Y.-H. Lin and C.-W. Nan, Solid State Ionics, 2011, 204–205, 41.
2. J.-M. Lee, T. Kim, S.-W. Baek, Y. Aihara, Y. Park, Y.-I. Kim and S.-G. Doo, Solid State Ionics, 2014, 258, 13.
3. J. Kim, Y. Yoon, M. Eom and D. Shin, Solid State Ionics, 2012, 225, 626.
4. S. Ito, M. Nakakita, Y. Aihara, T. Uehara and N. Machida, J. Power Sources, 2014, 271, 342.
5. S. Yubuchi, S. Teragawa, K. Aso, K. Tadanag, A. Hayashi and M. Tatsumisago, J. Power Sources, 2015, 293, 941.
6. K.-M. Kim, D.-O. Shin and Y.-G. Lee, Electrochim. Acta, 2015, 176, 1364.
7. E.-C. Bucharsky, K.-G. Schell, A. Hintennach and M.-J. Hoffmann, Solid State Ionics, 2015, 274, 77–82.
8. K. Arbi, W. Bucheli, R. Jiménez and J. Sanz, J. Eur. Ceram. Soc., 2015, 35, 1477.
9. C.-W. Ahn, J.-J. Choi, J. Ryu, B.-D. Hahn, J.-W. Kim, W.-H. Yoon, J.-H. Choi, J.-S. Lee and D.-S. Park, J. Power Sources, 2014, 272, 554.
10. H. Muramatsu, A. Hayashi, T. Ohtomo, S. Hama and M. Tatsumisago, Solid State Ionics, 2011, 182, 116.
11. T. Ohtomo, A. Hayashi, M. Tatsumisago and K. Kawamoto, J. Mater. Sci., 2013, 48, 4137.
12. P. Knauth, Solid State Ionics, 2009, 180, 911.
13. M. Kotobuki, Y. Suzuki, H. Munakata, K. Kanamura and T. Yoshida, Electrochim. Acta, 2011, 56, 1023.
14. J.-M. Lee, S.-H. Kim, Y. Tak and Y.-S. Yoon, J. Power Sources, 2006, 163, 173.
15. S. Iwasaki, T. Hamanaka, T. Yamakawa, W.-C. West, K. Yamamoto, M. Motoyama, T. Hirayama and Y. Iriyama, J. Power Sources, 2014, 272, 1086.
16. T. Kato, T. Hamanaka, K. Yamamoto, T. Hirayama, F. Sagane, M. Motoyama and Y. Iriyama, J. Power Sources, 2014, 260, 292.
17. Y. Lin, R. Ran, Y. Zheng, Z.-P. Shao, W.-Q. Jin, N.-P. Xu and J. Ahn, J. Power Sources, 2008, 180, 15.
18. G. Chen, H.-X. You, Y. Kasai, H. Sato and A. Abudula, J. Alloys Compd., 2011, 509, 5159.
19. X.-B. Zhu, Z. Lü, B. Wei, X.-Q. Huang, Y.-H. Zhang and W.-H. Su, J. Power Sources, 2011, 196, 729.
20. Z.-H. Bi, B.-L. Yi, Z.-W. Wang, Y.-L. Dong, H.-J. Wu, Y.-C. She and M.-J. Cheng, Electrochem. Solid-State Lett., 2004, 7, A105.
21. C.-R. Mariappan, C. Yada, F. Rosciano and B. Roling, J. Power Sources, 2011, 196, 6456.
22. M. Kotobuki and M. Koishi, Ceram. Int., 2015, 41, 8562.
23. C.-J. Leo, B.-V.-R. Chowdari, G.-V.-S. Rao and J.-L. Souquet, Mater. Res. Bull., 2002, 37, 1419.
24. X.-X. Xu, Z.-Y. Wen, X.-W. Wu, X.-L. Yang and Z.-H. Gu, J. Am. Ceram. Soc., 2007, 90, 2802.
25. J.-K. Feng, B.-G. Yan, J.-C. Liu, M.-O. Lai and L. Li, Mater. Technol., 2013, 28, 276.
26. F.-R. Ma, E.-Q. Zhao, S.-Y. Zhu, W.-C. Yan, D.-Y. Sun, Y.-C. Jin and C.-W. Nan, Solid State Ionics, 2016, 295, 7.
27. F.-M. Du, N. Zhao, Y.-Q. Li, C. Chen, Z.-W. Liu and X.-X. Guo, J. Power Sources, 2015, 300, 24.
28. J. Fu, J. Mater. Sci., 1998, 33, 1549.
29. L.-Z. Huang, Z.-Y. Wen, M.-F. Wu, X.-W. Wu, Y. Liu and X.-Y. Wang, J. Power Sources, 2011, 196, 6943.
30. H. Morimoto, M. Hirukawa, A. Matsumoto, T. Kurahayashi, N. Ito and S. Tobishima, Electrochemistry, 2014, 82, 870.
31. C. Davis and J.-C. Nino, J. Am. Ceram. Soc., 2015, 95, 2422.
32. S. Duluard, A. Paillassa, L. Puech, P. Vinatier, V. Turq, P. Rozier, P. Lenormand and P.-L. Taberna, J. Eur. Ceram. Soc., 2013, 33, 1145.
33. B.-G. Yan, Y.-Q. Zhu, F. Pan, J.-C. Liu and L. Lu, Solid State Ionics, 2015, 278, 65.

---

Footnote

† Electronic supplementary information (ESI) available: Details of the typical Nyquist plots of LATP and LAGP electrolytes measured at 25 °C, the complex impedance spectra of LATP and LAGP samples measured in the range of 303 K to 378 K, and the Arrhenius plots for the relationship between the total electrical conductivity and the inverse temperature are included here. See DOI: 10.1039/c6ra19415j

---