- Email: [email protected]
Fabrication and charge-discharge reaction of all solid-state lithium battery using Li4-2xGe1-xSxO4 electrolyte
Solid State Ionics 326 (2018) 52–57
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Fabrication and charge-discharge reaction of all solid-state lithium battery using Li4-2xGe1-xSxO4 electrolyte ⁎
T
⁎
Sou Taminato , Toyoki Okumura , Tomonari Takeuchi, Hironori Kobayashi National Institute of Advanced Industrial Science and Technology, 1-8-31 Midorigaoka, Ikeda, Osaka 563-8577, Japan
A R T I C LE I N FO
A B S T R A C T
Keywords: All solid-state battery Oxide electrolyte LISICON Lithium batteries
Bulk-type solid-state batteries using a LIthium Super Ionic CONductor (LISICON)-based oxide electrolyte, Li4were assembled by spark plasma sintering (SPS) and electric furnace sintering (FS), and their charge-discharge performances were investigated. Li4-2xGe1-xSxO4, which has a γ-Li3PO4-type structure, was synthesized by conventional solid-state reaction. The total ionic conductivity of the Li3.6Ge0.8S0.2O4 sample sintered at 600 °C by SPS was 2 × 10−5 S cm−1 at room temperature, which is comparable to the bulk conductivity of the material sintered at 800–1100 °C by FS. No impurity peaks were observed in the X-ray diffraction patterns of the LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 mixture even after high-temperature sintering at 900 °C by FS. The solid-state cells of laminated LiNi1/3Mn1/3Co1/3O2 cathode and Li3.6Ge0.8S0.2O4 electrolyte co-sintered by SPS at 600 °C and FS at 900 °C exhibited first discharge capacities of 130 and 92 mAh g−1, respectively, at 60 °C. These experimental results confirm that the LISICON-based material, Li4-2xGe1-xSxO4, exhibits thermal compatibility to the active material and relatively high lithium ion conductivity, which lead to the high chargedischarge performance of the bulk-type all-solid-state cell. 2xGe1-xSxO4,
1. Introduction All-solid-state lithium batteries with a non-flammable solid electrolyte have recently attracted significant attention owing to their high power and energy density and improved safety compared with the current battery system that uses a highly reactive organic electrolyte [1–3]. An electrode-electrolyte composite electrode usually shows large contact resistance for ion transfer at the interface. Sulfide electrolytes are extensively studied because their high ionic conductivity and suitable formability for the construction of the interface enable the cell to achieve a high charge-discharge performance [1,2,4,5]. On the other hand, because oxide electrolytes have lower moisture sensitivity and toxicity than sulfide electrolytes, they are anticipated to improve the reliability of the battery. However, it is a challenge to use these electrolytes to construct a well-formed electrode-electrolyte composite electrode mainly owing to their poor deformability at room temperature. Although a sintering process is necessary to create good contact between the electrode and electrolyte, high-temperature sintering often leads to the formation of resistive interfacial phases [6–8]. Either lowtemperature sintering or the use of an oxide electrolyte with high thermal stability against the electrode during the sintering process is an important factor in providing a reversible charge-discharge reaction during battery operation. Spark plasma sintering (SPS) is a well-known ⁎
technique that enables low-temperature or rapid sintering, and it is used to produce dense ceramics [7,9,10]. Although this technique has been applied to achieve good contact between the electrode and solid electrolyte within a short period, using it for the co-sintering of a LiTi2(PO4)3 and LiCoO2 mixture resulted in the formation of interfacial impurity [7]. Recently, we have reported that a well-defined interface between LiCoO2 and the Li2CO3-Li3BO3 electrolyte was achieved without formation of any impurity phases by co-sintering the materials by SPS. This resulted in the reversible lithium (de-)intercalation reactions of the LiCoO2 cathode material for the bulk-type solid-state cell [10]. The Li2CO3-Li3BO3 electrolyte has high thermal stability against the electrode material at the co-sintering temperature; however, it exhibits low conductivity at 30 °C (6.5 × 10−7 S cm−1) [11], which could limit the reversible capacity of the cell. An oxide electrolyte with high ionic conductivity is therefore desirable to further improve the battery performance of an all-solid-state cell. LIthium Super Ionic CONductor (LISICON)-related materials with a γ-Li3PO4 type structure in the germanium system have conductivities of approximately 10−5 S cm−1 at room temperature and form a solid solution with many elements for a wide range of composition [12–16]. In particular, Li4(Ge, Ti)O4-Li3(As, V)O4 and Li4GeO4-Li2SO4 have conductivities of 2–4 × 10−5 S cm−1 at room temperature [12,16]. In a preliminary experiment, LISICON materials exhibited high thermal
Corresponding authors. E-mail addresses: [email protected] (S. Taminato), [email protected] (T. Okumura).
https://doi.org/10.1016/j.ssi.2018.09.011 Received 30 June 2018; Received in revised form 17 September 2018; Accepted 17 September 2018 0167-2738/ © 2018 Elsevier B.V. All rights reserved.
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
placed in a 10-mm stainless steel die, on top of which was placed 10 mg of the composite electrode powder. The powders were pressed together under a pressure of 800 MPa, and then sintered at 800–900 °C for 2 h with oxygen flowing through the electric furnace. After sintering, a gold current collector was deposited onto the composite electrode by DC sputtering. The same amounts of the composite electrode and electrolyte powders were used for SPS. The composite electrode powder and gold current collector were placed onto the Li3.6Ge0.8S0.2O4 electrolyte powder in a 10-mm carbon die, which was then heated at 600 °C for 1 min under a pressure of 30 MPa in Ar atmosphere in the SPS instrument. Lithium foil was used as a counter electrode. A polyethylene oxide (PEO)-based polymer electrolyte film (LiTFSA/EO = 0.06, Osaka Soda, Japan) was placed between the lithium foil and electrolyte side of the composite electrode/electrolyte pellet to reduce the interfacial resistance upon adhesion [10]. The chargedischarge characteristics of the all-solid-state cells assembled by FS and SPS were examined in the range of 2.0–4.2 V with a constant current of 32 μA cm−2 (4.2 mA g−1 of LiNi1/3Mn1/3Co1/3O2) at 60 °C using a charge-discharge testing apparatus (BST2004, Nagano, Japan). The cross-sectional microstructures of the composite electrode/electrolyte sintered by SPS and FS were observed using scanning electron microscopy (SEM, JSM-6510LA, JEOL Ltd., Japan). The cross-sections of the samples were prepared by polishing using Ar ion-milling (IB-09020CP, JEOL Ltd., Japan).
stability against layered rock salt electrodes, in contrast to well-known oxide electrolytes with high ionic conductivity, such as NASICON and garnet-type structure [7,8]. Thus, these materials can potentially be used as electrolytes for a bulk-type all-solid-state battery to improve its limited battery performance. In this study, we focused on the LISICONbased material, Li4GeO4-Li2SO4 (Li4-2xGe1-xSxO4), which exhibits high ionic conductivity and consists of typical metal cations acting as the solid electrolyte. Its thermal compatibility to the layered rock salt cathode, LiNi1/3Mn1/3Co1/3O2, during the electric furnace sintering (FS) and SPS processes was examined. Using Li3.6Ge0.8S0.2O4 as the electrolyte, the charge-discharge characteristics of all-solid-state cells assembled by these sintering processes were investigated. 2. Experimental During the synthesis of the electrolyte materials, oxygen flow was maintained to prevent Li2CO3 formation on the prepared powder as much as possible [17]. Li4GeO4 was prepared using the starting materials, LiOH·H2O (99.0% purity, Kojundo Chemical Laboratory, Japan) and GeO2 (99.0% purity, Kojundo Chemical Laboratory, Japan). Appropriate amounts of the compounds were weighed to obtain a molar ratio of 4:1, and then the mixture was ground in an agate mortar. The powder mixture was fired at 600 °C for 12 h and subsequently pelletized and sintered at 700 °C for 12 h. Li4-2xGe1-xSxO4 (x = 0.1–0.4) was synthesized by sintering the powder mixture of Li4GeO4 and Li2SO4. Appropriate amounts of ground Li4GeO4 and Li2SO4 (99.0% purity, Kojundo Chemical Laboratory, Japan) were weighed to obtain the desired molar ratio, and then the compounds were mixed in an agate mortar within an Ar-filled glove box. The final mixture was pressed into pellets and heated at 900 °C for 12 h. Phase identification of the Li4-2xGe1-xSxO4 samples was carried out by powder X-ray diffraction (XRD) using an Xray diffractometer with Cu Kα radiation (Miniflex600, Rigaku, Japan). Lattice parameters were refined using the Rietveld refinement program, RIETAN-FP [18]. The Li4-2xGe1-xSxO4 samples were crushed in a ZrO2 pot with five 5-mm ZrO2 disks using vibration milling (MC-4A, Ito Seisakusho, Japan). The crushed samples were sandwiched between the Li-ion-blocking gold electrodes that will be used in the AC impedance measurement, and then placed in a 10-mm carbon die. To prepare the samples by SPS, the die was heated to 600 °C at a rate of ≈50 °C min−1 with a DC sawtooth-pulsed electric current under a pressure of 30 MPa (pulse length: 2.5 ms, number of consecutive pulses: 14 times, interval after consecutive pulses: 5 ms) in the SPS instrument (SPS-515S, Sumitomo Coal Mining, Japan). The After reaching the desired temperature, the conditions were kept for 1 min, after which the applied current was stopped, pressure was released, and the sample was cooled to room temperature. The ionic conductivity of the sintered samples was measured by the AC impedance method under Ar atmosphere at room temperature. An alternating voltage with an amplitude of 50 mV was applied in the frequency range between 1 and 30 MHz using the 1260 Frequency Response Analyzer (Solartron Analytical, UK). The compatibility between the cathode and electrolyte materials sintered by FS and SPS was examined by XRD. LiNi1/3Mn1/3Co1/3O2 (average particle size = 10.0 μm, Nippon Chemical Industrial, Japan) and Li3.6Ge0.8S0.2O4 were mixed (1:1 w/w%) in an agate mortar. The mixture was pressed under a pressure of 800 MPa in a 10-mm stainless steel die to make a pellet, which was then sintered at 900 °C for 2 h with oxygen flowing through the electric furnace. To prepare the samples by SPS, the mixture was sandwiched between the titanium electrodes, placed in a 10-mm carbon die, and sintered at 600 °C under a pressure 30 MPa for 1 min in Ar atmosphere. A composite electrode powder consisting of 60 wt% LiNi1/3Mn1/ 3Co1/3O2 and 40 wt% Li3.6Ge0.8S0.2O4 (i.e., Li4-2xGe1-xSxO4, x = 0.2) electrolyte was prepared by mixing the materials in an agate mortar. A disk-shaped half-cell pellet containing the composite electrode/electrolyte layer was prepared by SPS and FS for the charge-discharge measurement. The Li3.6Ge0.8S0.2O4 electrolyte powder (30 mg) was
3. Results and discussion Fig. 1a shows the powder XRD patterns of Li4-2xGe1-xSxO4 samples with a composition within the range of 0 ≤ x ≤ 0.4. The sample with x = 0 exhibits the diffraction peaks attributable to the Li4GeO4 phase having the Cmcm space group [12]. On the other hand, the samples with x ≠ 0 had diffraction patterns that mainly correspond to the LISICON phase having a γ-Li3PO4-type structure (space group Pnma) [12]. The peaks derived from the secondary phase are not observed in samples with x = 0.1–0.3. These results confirm the formation of the LISICON phase upon substitution of Ge by S in Li4GeO4. The peaks gradually shift to higher angle with increasing x, indicating the formation of solid solution. The sample with x = 0.4 shows a two-phase character due to the LISICON phase and β-Li2SO4. The lattice parameters of Li42xGe1-xSxO4 (x = 0.1–0.4), refined using the RIETAN-FP program, are shown in Fig. 1b. The a and c axes linearly decrease, while b slightly increases, with increasing x. The cell volume decreases linearly with increasing x because of the replacement of Ge4+ (r = 0.39 Å) by the smaller S6+ cation (r = 0.12 Å) in the structure [12,19]. The composition dependence of the ionic conductivity at room temperature was determined for the Li4-2xGe1-xSxO4 (x = 0.1–0.4) samples sintered at 600 °C by SPS, and the Nyquist plots are shown in Fig. 2a. The spectra contain a semicircle and spike in the high- and lowfrequency regions, respectively. The semicircle likely corresponds to the bulk resistance of Li4-2xGe1-xSxO4, while the spike is due to charge polarization at the blocking electrode. Note that although grain-boundary resistance is often observed for oxide electrolyte pellets sintered by FS, this resistance is minimized after SPS [17]. The resistance was determined from the intercept of the spike with the real axis. The ionic conductivities of the Li4-2xGe1-xSxO4 (x = 0.1–0.4) samples, calculated from the bulk resistance, are summarized in Fig. 2b. The conductivity increases from x = 0.1 to 0.2, but decreases from x = 0.2 to 0.4 because of an optimization in the carrier concentration and cooperative interactions between the mobile lithium ions [12]. In this study, the highest conductivity, 2 × 10−5 S cm−1, was obtained for Li3.6Ge0.8S0.2O4, which is comparable to the bulk conductivity of the material sintered at 800–1100 °C by FS [12]. These results demonstrate that Li4-2xGe1-xSxO4 was successfully sintered by SPS at a lower temperature than that used in FS, and the grain-boundary resistance was reduced by SPS. Therefore, Li3.6Ge0.8S0.2O4 was used as the separator layer and electrolyte material in the composite electrode of the solid-state cell fabricated in this study. 53
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 2. (a) Nyquist plots and (b) conductivities for the synthesized Li4-2xGe1(x = 0.1–0.4) sintered at 600 °C using SPS process. All impedance measurement conducted at room temperature.
xSxO4
regions corresponding to these reactions [20] are apparent from 3.8 to 4.2 V in the charge process and 3.6 to 4.0 V in the discharge process. An overpotential of ca. 200 mV is observed in the cell. The cell exhibits first charge and discharge capacities of 101 and 50 mAh g−1, respectively, with an irreversible capacity of 51 mAh g−1. At a sintering temperature of 900 °C, the cell exhibits a smaller overpotential than that at 800 °C and larger charge-discharge capacities of 143 and 92 mAh g−1, as shown in Fig. 4b. The enhancement of the cell performance may be attributed to the stronger contact between LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4. Fig. 4c shows the charge-discharge curves for the allsolid-state cell assembled by SPS. In contrast to the cell assembled by FS, the slope regions are apparent around 3.6 V in both the charge and discharge processes. The cell exhibits a first charge capacity of 163 mAh g−1, followed by reversible and irreversible capacities of 130 and 33 mAh g−1, respectively. The discharge capacity decreases from 130 to 125 mAh g−1 during the initial 5 cycles. The initial cycle performance is higher than that of an all-solid-state cell assembled by SPS using Li2.2C0.8B0.2O3 as electrolyte [10]. The charge-discharge capacity of the cell assembled by SPS is higher than that of the cell assembled by FS, although no impurity phase was formed between LiNi1/3Mn1/3Co1/ 3O2 and Li3.6Ge0.8S0.2O4 during both processes. These results confirm that both solid-state cells exhibit the reversible charge-discharge
Fig. 1. (a) Powder X-ray diffraction patterns, and (b) lattice parameters a, b, c and cell volume V calculated as orthorhombic cell for the synthesized Li4-2xGe1xSxO4 (x = 0–0.4).
Fig. 3 shows the XRD patterns of the LiNi1/3Mn1/3Co1/3O2Li3.6Ge0.8S0.2O4 mixtures sintered at 900 °C by FS and 600 °C by SPS. The only peaks observed in both XRD patterns are the original peaks derived from LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4. Most oxide electrolytes react with the layered rock salt cathode during the sintering process and forms impurity phases, resulting in the degradation of the charge-discharge characteristics of the all-solid-state cell [6–8]. In contrast, Li3.6Ge0.8S0.2O4 does not form impurities during the sintering process with LiNi1/3Mn1/3Co1/3O2, whether the process is the lowtemperature SPS or high-temperature FS. This result reveals that Li3.6Ge0.8S0.2O4 can be used to fabricate a composite electrode with LiNi1/3Mn1/3Co1/3O2 by FS. Fig. 4a shows the charge-discharge curves for the all-solid-state cell assembled by FS at 800 °C. The electrochemical lithium (de-)intercalation reactions of the LiNi1/3Mn1/3Co1/3O2 cathode are observed in the all-solid-state cell using Li3.6Ge0.8S0.2O4 as electrolyte. The slope 54
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 3. Powder X-ray diffraction patterns of (a) LiNi1/3Mn1/3Co1/3O2, (b) Li3.6Ge0.8S0.2O4, (c) LiNi1/3Mn1/3Co1/3O2/Li3.6Ge0.8S0.2O4 mixture sintered at 900 °C by FS and (d) 600 °C by SPS process.
reaction, although the cell fabricated using SPS delivered higher chargedischarge characteristics than that using FS. Fig. 5 shows the SEM images of the cross-sectional microstructures of the composite electrode/electrolyte sintered at 800 and 900 °C by FS, and 600 °C by SPS. Although the ratio of electrode material/electrolyte sintered was the same in each process, FS yielded a thicker composite electrode layer (ca. 50 μm) than SPS (ca. 25 μm). The calculated porosity of the composite electrode sintered by FS was 39% because only a little uniaxial pressure was applied to the electrode/electrolyte layer during the sintering process. However, the porosity of the sample sintered by SPS could not be calculated correctly because the thickness of the electrode was smaller than that estimated without the pores. A part of the sample might have been peeled off when the sample was removed from the electrode after the SPS experiment to prepare it for SEM. Therefore, the electrode thickness during the battery experiment would be larger. For the composite electrode layer sintered at 800 °C by FS, Fig. 5a shows that LiNi1/3Mn1/3Co1/3O2 particles, with sizes of 1–10 μm, are dispersed with connection between each other, and there are voids between the LiNi1/3Mn1/3Co1/3O2/Li3.6Ge0.8S0.2O4 and Li3.6Ge0.8S0.2O4/Li3.6Ge0.8S0.2O4 phases. Upon increasing the sintering temperature to 900 °C, the domains of the Li3.6Ge0.8S0.2O4 phase become larger and apparent contact between the active material and electrolyte is observed because of the progress in the sintering process (Fig. 5b). In the microstructure of the composite electrode layer sintered by SPS, the LiNi1/3Mn1/3Co1/3O2 particles are dispersed similarly as in FS, and the Li3.6Ge0.8S0.2O4 electrolyte is packed densely. This results in a well-defined interface between the LiNi1/3Mn1/3Co1/3O2/ Li3.6Ge0.8S0.2O4, and Li3.6Ge0.8S0.2O4/Li3.6Ge0.8S0.2O4 phases. The insufficiently connected interfaces of the active material/electrolyte and electrolyte/electrolyte phases in the composite electrode layer sintered by FS could result in a large internal resistance and is one of the causes of the poor charge-discharge performance. Utilization of a LISICONbased material, Li4-2xGe1-xSxO4, that have a γ-Li3PO4-type structure as the solid electrolyte for an all-solid-state battery suppresses the reaction with the active material during the sintering process. Moreover, it permits the fabrication of a cell with a reversible charge-discharge reaction using SPS, which provides dense sintering at a relatively low temperature. The hot-pressing technique has also been reported to
Fig. 4. Charge-discharge curves of the LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 composite cathode/Li3.6Ge0.8S0.2O4 electrolyte/PEO film/lithium foil cell. The layered composite cathode and electrolyte component was sintered at (a) 800 °C and (b) 900 °C by FS process, and (c) 600 °C by SPS process, respectively.
result in dense sintering for oxide ceramics, although it requires a higher temperature for sintering than SPS [21]. Based on our experimental results, Li4-2xGe1-xSxO4 shows high thermal compatibility to LiNi1/3Mn1/3Co1/3O2 during the sintering process. Thus, a high-temperature and high-pressure sintering process such as the hot-pressing method can be applied in the fabrication of an all-solid-state cell with reversible charge-discharge reaction if Li4-2xGe1-xSxO4 is used as the electrolyte.
4. Conclusion The use of the LISICON-based material, Li4-2xGe1-xSxO4, as electrolyte suppressed the reaction with LiNi1/3Mn1/3Co1/3O2 during the cosintering process by FS at temperatures as high as 900 °C and led to the reversible charge-discharge reaction in the solid-state cell. A solid solution was formed for samples with a composition x between 0.1 and 0.4. The Li3.6Ge0.8S0.2O4 sample sintered at 600 °C by SPS exhibited an ionic conductivity of 2 × 10−5 S cm−1 at room temperature, which is 55
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 5. Cross sectional SEM images of LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 composite cathode/Li3.6Ge0.8S0.2O4 electrolyte layer sintered at (a) 800 °C and (c) 900 °C by FS, and (e) 600 °C by SPS processes. The expanded images in the composite cathode sintered at (b) 800 °C and (d) 900 °C by FS, and (f) 600 °C by SPS methods.
to thank Ms. Wagou, Mr. Watanabe and Dr. Yamamoto for helping the synthesis, SPS and SEM experiments.
mainly from bulk contribution. This indicates that the resistance derived from grain boundary was decreased by SPS. Although no phase formation was observed in the mixture of LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4 during high-temperature sintering by FS, the assembled cell exhibited limited discharge capacities of 50 and 92 mAh g−1 at 800 and 900 °C, respectively, because of the poor connection between the active material/electrolyte and electrolyte/electrolyte interfaces in the composite layer. In contrast, the cell assembled by SPS had well-connected interfaces and exhibited a larger discharge capacity of 130 mAh g−1 and better initial cycle performance.
References [1] Y. Kato, S. Hori, T. Saito, K. Suzuki, M. Hirayama, A. Mitsui, M. Yonemura, H. Iba, R. Kanno, Nat. Energy 1 (16030) (2016) 1–7. [2] Y. Kato, K. Kawamoto, R. Kanno, M. Hirayama, Electrochemistry 80 (10) (2012) 749–751. [3] A.L. Robinson, J. Janek, MRS Bull. 39 (12) (2014) 1046–1047. [4] A. Sakuda, A. Hayashi, M. Tatsumisago, Sci. Rep. 3 (2261) (2013) 1–5. [5] M. Nose, A. Kato, A. Sakuda, A. Hayashi, M. Tatsumisago, J. Mater. Chem. A 3 (2015) 22061–22065. [6] J. Xie, N. Imanishi, T. Zhang, A. Hirano, Y. Takeda, O. Yamamoto, J. Power Sources 189 (2009) 365–370. [7] Y. Kobayashi, T. Takeuchi, M. Tabuchi, K. Ado, K. Kageyama, J. Power Sources 81–82 (1999) 853–858. [8] K. Park, B.-C. Yu, J.-W. Jung, Y. Li, W. Zhou, H. Gao, S. Son, J.B. Goodenough, Chem. Mater. 28 (21) (2016) 8051–8059. [9] A. Mei, Q.-H. Jiang, Y.-H. Lin, C.-W. Nan, J. Alloys Compd. 486 (2008) 871–875. [10] T. Okumura, T. Takeuchi, H. Kobayashi, J. Ceram. Soc. Jpn. 125 (4) (2017)
Acknowledgements This work was financially supported by the Advanced Low Carbon Technology Research and Development Program of the Japan Science and Technology Agency for Specially Promoted Research for Innovative Next Generation Batteries (JST-ALCA SPRING). The authors would like 56
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
[17] T. Okumura, S. Taminato, T. Takeuchi, H. Kobayashi, ACS Appl. Energy Mater. (2018) (submitted). [18] F. Izumi, K. Momma, Solid State Phenom. 130 (2007) 15–20. [19] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751–767. [20] N. Yabuuchi, Y. Makimura, T. Ohzuku, J. Electrochem. Soc. 154 (4) (2007) A314–A321. [21] S.B. Li, C.B. Wang, D.Q. Zhou, H.X. Liu, L. Li, Q. Shen, L.M. Zhang, Ceram. Int. 44 (1) (2018) 550–555.
276–280. [11] T. Okumura, T. Takeuchi, H. Kobayashi, Solid State Ionics 188 (2016) 248–252. [12] M.A.K.L. Dissanayake, R.P. Gunawardane, A.R. West, G.K.R. Senadeera, P.W.S.K. Bandaranayake, M.A. Careem, Solid State Ionics 62 (3–4) (1993) 217–223. [13] H.Y.-P. Hong, Mater. Res. Bull. 13 (2) (1978) 117–124. [14] J.G. Kamphorst, E.E. Hellstrom, Solid State Ionics 1 (3–4) (1980) 187–197. [15] J. Kuwano, A.R. West, Mater. Res. Bull. 15 (11) (1980) 1661–1667. [16] A.R. Rodger, J. Kuwano, A.R. West, Solid State Ionics 15 (3) (1985) 185–198.
57
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Fabrication and charge-discharge reaction of all solid-state lithium battery using Li4-2xGe1-xSxO4 electrolyte ⁎
T
⁎
Sou Taminato , Toyoki Okumura , Tomonari Takeuchi, Hironori Kobayashi National Institute of Advanced Industrial Science and Technology, 1-8-31 Midorigaoka, Ikeda, Osaka 563-8577, Japan
A R T I C LE I N FO
A B S T R A C T
Keywords: All solid-state battery Oxide electrolyte LISICON Lithium batteries
Bulk-type solid-state batteries using a LIthium Super Ionic CONductor (LISICON)-based oxide electrolyte, Li4were assembled by spark plasma sintering (SPS) and electric furnace sintering (FS), and their charge-discharge performances were investigated. Li4-2xGe1-xSxO4, which has a γ-Li3PO4-type structure, was synthesized by conventional solid-state reaction. The total ionic conductivity of the Li3.6Ge0.8S0.2O4 sample sintered at 600 °C by SPS was 2 × 10−5 S cm−1 at room temperature, which is comparable to the bulk conductivity of the material sintered at 800–1100 °C by FS. No impurity peaks were observed in the X-ray diffraction patterns of the LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 mixture even after high-temperature sintering at 900 °C by FS. The solid-state cells of laminated LiNi1/3Mn1/3Co1/3O2 cathode and Li3.6Ge0.8S0.2O4 electrolyte co-sintered by SPS at 600 °C and FS at 900 °C exhibited first discharge capacities of 130 and 92 mAh g−1, respectively, at 60 °C. These experimental results confirm that the LISICON-based material, Li4-2xGe1-xSxO4, exhibits thermal compatibility to the active material and relatively high lithium ion conductivity, which lead to the high chargedischarge performance of the bulk-type all-solid-state cell. 2xGe1-xSxO4,
1. Introduction All-solid-state lithium batteries with a non-flammable solid electrolyte have recently attracted significant attention owing to their high power and energy density and improved safety compared with the current battery system that uses a highly reactive organic electrolyte [1–3]. An electrode-electrolyte composite electrode usually shows large contact resistance for ion transfer at the interface. Sulfide electrolytes are extensively studied because their high ionic conductivity and suitable formability for the construction of the interface enable the cell to achieve a high charge-discharge performance [1,2,4,5]. On the other hand, because oxide electrolytes have lower moisture sensitivity and toxicity than sulfide electrolytes, they are anticipated to improve the reliability of the battery. However, it is a challenge to use these electrolytes to construct a well-formed electrode-electrolyte composite electrode mainly owing to their poor deformability at room temperature. Although a sintering process is necessary to create good contact between the electrode and electrolyte, high-temperature sintering often leads to the formation of resistive interfacial phases [6–8]. Either lowtemperature sintering or the use of an oxide electrolyte with high thermal stability against the electrode during the sintering process is an important factor in providing a reversible charge-discharge reaction during battery operation. Spark plasma sintering (SPS) is a well-known ⁎
technique that enables low-temperature or rapid sintering, and it is used to produce dense ceramics [7,9,10]. Although this technique has been applied to achieve good contact between the electrode and solid electrolyte within a short period, using it for the co-sintering of a LiTi2(PO4)3 and LiCoO2 mixture resulted in the formation of interfacial impurity [7]. Recently, we have reported that a well-defined interface between LiCoO2 and the Li2CO3-Li3BO3 electrolyte was achieved without formation of any impurity phases by co-sintering the materials by SPS. This resulted in the reversible lithium (de-)intercalation reactions of the LiCoO2 cathode material for the bulk-type solid-state cell [10]. The Li2CO3-Li3BO3 electrolyte has high thermal stability against the electrode material at the co-sintering temperature; however, it exhibits low conductivity at 30 °C (6.5 × 10−7 S cm−1) [11], which could limit the reversible capacity of the cell. An oxide electrolyte with high ionic conductivity is therefore desirable to further improve the battery performance of an all-solid-state cell. LIthium Super Ionic CONductor (LISICON)-related materials with a γ-Li3PO4 type structure in the germanium system have conductivities of approximately 10−5 S cm−1 at room temperature and form a solid solution with many elements for a wide range of composition [12–16]. In particular, Li4(Ge, Ti)O4-Li3(As, V)O4 and Li4GeO4-Li2SO4 have conductivities of 2–4 × 10−5 S cm−1 at room temperature [12,16]. In a preliminary experiment, LISICON materials exhibited high thermal
Corresponding authors. E-mail addresses: [email protected] (S. Taminato), [email protected] (T. Okumura).
https://doi.org/10.1016/j.ssi.2018.09.011 Received 30 June 2018; Received in revised form 17 September 2018; Accepted 17 September 2018 0167-2738/ © 2018 Elsevier B.V. All rights reserved.
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
placed in a 10-mm stainless steel die, on top of which was placed 10 mg of the composite electrode powder. The powders were pressed together under a pressure of 800 MPa, and then sintered at 800–900 °C for 2 h with oxygen flowing through the electric furnace. After sintering, a gold current collector was deposited onto the composite electrode by DC sputtering. The same amounts of the composite electrode and electrolyte powders were used for SPS. The composite electrode powder and gold current collector were placed onto the Li3.6Ge0.8S0.2O4 electrolyte powder in a 10-mm carbon die, which was then heated at 600 °C for 1 min under a pressure of 30 MPa in Ar atmosphere in the SPS instrument. Lithium foil was used as a counter electrode. A polyethylene oxide (PEO)-based polymer electrolyte film (LiTFSA/EO = 0.06, Osaka Soda, Japan) was placed between the lithium foil and electrolyte side of the composite electrode/electrolyte pellet to reduce the interfacial resistance upon adhesion [10]. The chargedischarge characteristics of the all-solid-state cells assembled by FS and SPS were examined in the range of 2.0–4.2 V with a constant current of 32 μA cm−2 (4.2 mA g−1 of LiNi1/3Mn1/3Co1/3O2) at 60 °C using a charge-discharge testing apparatus (BST2004, Nagano, Japan). The cross-sectional microstructures of the composite electrode/electrolyte sintered by SPS and FS were observed using scanning electron microscopy (SEM, JSM-6510LA, JEOL Ltd., Japan). The cross-sections of the samples were prepared by polishing using Ar ion-milling (IB-09020CP, JEOL Ltd., Japan).
stability against layered rock salt electrodes, in contrast to well-known oxide electrolytes with high ionic conductivity, such as NASICON and garnet-type structure [7,8]. Thus, these materials can potentially be used as electrolytes for a bulk-type all-solid-state battery to improve its limited battery performance. In this study, we focused on the LISICONbased material, Li4GeO4-Li2SO4 (Li4-2xGe1-xSxO4), which exhibits high ionic conductivity and consists of typical metal cations acting as the solid electrolyte. Its thermal compatibility to the layered rock salt cathode, LiNi1/3Mn1/3Co1/3O2, during the electric furnace sintering (FS) and SPS processes was examined. Using Li3.6Ge0.8S0.2O4 as the electrolyte, the charge-discharge characteristics of all-solid-state cells assembled by these sintering processes were investigated. 2. Experimental During the synthesis of the electrolyte materials, oxygen flow was maintained to prevent Li2CO3 formation on the prepared powder as much as possible [17]. Li4GeO4 was prepared using the starting materials, LiOH·H2O (99.0% purity, Kojundo Chemical Laboratory, Japan) and GeO2 (99.0% purity, Kojundo Chemical Laboratory, Japan). Appropriate amounts of the compounds were weighed to obtain a molar ratio of 4:1, and then the mixture was ground in an agate mortar. The powder mixture was fired at 600 °C for 12 h and subsequently pelletized and sintered at 700 °C for 12 h. Li4-2xGe1-xSxO4 (x = 0.1–0.4) was synthesized by sintering the powder mixture of Li4GeO4 and Li2SO4. Appropriate amounts of ground Li4GeO4 and Li2SO4 (99.0% purity, Kojundo Chemical Laboratory, Japan) were weighed to obtain the desired molar ratio, and then the compounds were mixed in an agate mortar within an Ar-filled glove box. The final mixture was pressed into pellets and heated at 900 °C for 12 h. Phase identification of the Li4-2xGe1-xSxO4 samples was carried out by powder X-ray diffraction (XRD) using an Xray diffractometer with Cu Kα radiation (Miniflex600, Rigaku, Japan). Lattice parameters were refined using the Rietveld refinement program, RIETAN-FP [18]. The Li4-2xGe1-xSxO4 samples were crushed in a ZrO2 pot with five 5-mm ZrO2 disks using vibration milling (MC-4A, Ito Seisakusho, Japan). The crushed samples were sandwiched between the Li-ion-blocking gold electrodes that will be used in the AC impedance measurement, and then placed in a 10-mm carbon die. To prepare the samples by SPS, the die was heated to 600 °C at a rate of ≈50 °C min−1 with a DC sawtooth-pulsed electric current under a pressure of 30 MPa (pulse length: 2.5 ms, number of consecutive pulses: 14 times, interval after consecutive pulses: 5 ms) in the SPS instrument (SPS-515S, Sumitomo Coal Mining, Japan). The After reaching the desired temperature, the conditions were kept for 1 min, after which the applied current was stopped, pressure was released, and the sample was cooled to room temperature. The ionic conductivity of the sintered samples was measured by the AC impedance method under Ar atmosphere at room temperature. An alternating voltage with an amplitude of 50 mV was applied in the frequency range between 1 and 30 MHz using the 1260 Frequency Response Analyzer (Solartron Analytical, UK). The compatibility between the cathode and electrolyte materials sintered by FS and SPS was examined by XRD. LiNi1/3Mn1/3Co1/3O2 (average particle size = 10.0 μm, Nippon Chemical Industrial, Japan) and Li3.6Ge0.8S0.2O4 were mixed (1:1 w/w%) in an agate mortar. The mixture was pressed under a pressure of 800 MPa in a 10-mm stainless steel die to make a pellet, which was then sintered at 900 °C for 2 h with oxygen flowing through the electric furnace. To prepare the samples by SPS, the mixture was sandwiched between the titanium electrodes, placed in a 10-mm carbon die, and sintered at 600 °C under a pressure 30 MPa for 1 min in Ar atmosphere. A composite electrode powder consisting of 60 wt% LiNi1/3Mn1/ 3Co1/3O2 and 40 wt% Li3.6Ge0.8S0.2O4 (i.e., Li4-2xGe1-xSxO4, x = 0.2) electrolyte was prepared by mixing the materials in an agate mortar. A disk-shaped half-cell pellet containing the composite electrode/electrolyte layer was prepared by SPS and FS for the charge-discharge measurement. The Li3.6Ge0.8S0.2O4 electrolyte powder (30 mg) was
3. Results and discussion Fig. 1a shows the powder XRD patterns of Li4-2xGe1-xSxO4 samples with a composition within the range of 0 ≤ x ≤ 0.4. The sample with x = 0 exhibits the diffraction peaks attributable to the Li4GeO4 phase having the Cmcm space group [12]. On the other hand, the samples with x ≠ 0 had diffraction patterns that mainly correspond to the LISICON phase having a γ-Li3PO4-type structure (space group Pnma) [12]. The peaks derived from the secondary phase are not observed in samples with x = 0.1–0.3. These results confirm the formation of the LISICON phase upon substitution of Ge by S in Li4GeO4. The peaks gradually shift to higher angle with increasing x, indicating the formation of solid solution. The sample with x = 0.4 shows a two-phase character due to the LISICON phase and β-Li2SO4. The lattice parameters of Li42xGe1-xSxO4 (x = 0.1–0.4), refined using the RIETAN-FP program, are shown in Fig. 1b. The a and c axes linearly decrease, while b slightly increases, with increasing x. The cell volume decreases linearly with increasing x because of the replacement of Ge4+ (r = 0.39 Å) by the smaller S6+ cation (r = 0.12 Å) in the structure [12,19]. The composition dependence of the ionic conductivity at room temperature was determined for the Li4-2xGe1-xSxO4 (x = 0.1–0.4) samples sintered at 600 °C by SPS, and the Nyquist plots are shown in Fig. 2a. The spectra contain a semicircle and spike in the high- and lowfrequency regions, respectively. The semicircle likely corresponds to the bulk resistance of Li4-2xGe1-xSxO4, while the spike is due to charge polarization at the blocking electrode. Note that although grain-boundary resistance is often observed for oxide electrolyte pellets sintered by FS, this resistance is minimized after SPS [17]. The resistance was determined from the intercept of the spike with the real axis. The ionic conductivities of the Li4-2xGe1-xSxO4 (x = 0.1–0.4) samples, calculated from the bulk resistance, are summarized in Fig. 2b. The conductivity increases from x = 0.1 to 0.2, but decreases from x = 0.2 to 0.4 because of an optimization in the carrier concentration and cooperative interactions between the mobile lithium ions [12]. In this study, the highest conductivity, 2 × 10−5 S cm−1, was obtained for Li3.6Ge0.8S0.2O4, which is comparable to the bulk conductivity of the material sintered at 800–1100 °C by FS [12]. These results demonstrate that Li4-2xGe1-xSxO4 was successfully sintered by SPS at a lower temperature than that used in FS, and the grain-boundary resistance was reduced by SPS. Therefore, Li3.6Ge0.8S0.2O4 was used as the separator layer and electrolyte material in the composite electrode of the solid-state cell fabricated in this study. 53
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 2. (a) Nyquist plots and (b) conductivities for the synthesized Li4-2xGe1(x = 0.1–0.4) sintered at 600 °C using SPS process. All impedance measurement conducted at room temperature.
xSxO4
regions corresponding to these reactions [20] are apparent from 3.8 to 4.2 V in the charge process and 3.6 to 4.0 V in the discharge process. An overpotential of ca. 200 mV is observed in the cell. The cell exhibits first charge and discharge capacities of 101 and 50 mAh g−1, respectively, with an irreversible capacity of 51 mAh g−1. At a sintering temperature of 900 °C, the cell exhibits a smaller overpotential than that at 800 °C and larger charge-discharge capacities of 143 and 92 mAh g−1, as shown in Fig. 4b. The enhancement of the cell performance may be attributed to the stronger contact between LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4. Fig. 4c shows the charge-discharge curves for the allsolid-state cell assembled by SPS. In contrast to the cell assembled by FS, the slope regions are apparent around 3.6 V in both the charge and discharge processes. The cell exhibits a first charge capacity of 163 mAh g−1, followed by reversible and irreversible capacities of 130 and 33 mAh g−1, respectively. The discharge capacity decreases from 130 to 125 mAh g−1 during the initial 5 cycles. The initial cycle performance is higher than that of an all-solid-state cell assembled by SPS using Li2.2C0.8B0.2O3 as electrolyte [10]. The charge-discharge capacity of the cell assembled by SPS is higher than that of the cell assembled by FS, although no impurity phase was formed between LiNi1/3Mn1/3Co1/ 3O2 and Li3.6Ge0.8S0.2O4 during both processes. These results confirm that both solid-state cells exhibit the reversible charge-discharge
Fig. 1. (a) Powder X-ray diffraction patterns, and (b) lattice parameters a, b, c and cell volume V calculated as orthorhombic cell for the synthesized Li4-2xGe1xSxO4 (x = 0–0.4).
Fig. 3 shows the XRD patterns of the LiNi1/3Mn1/3Co1/3O2Li3.6Ge0.8S0.2O4 mixtures sintered at 900 °C by FS and 600 °C by SPS. The only peaks observed in both XRD patterns are the original peaks derived from LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4. Most oxide electrolytes react with the layered rock salt cathode during the sintering process and forms impurity phases, resulting in the degradation of the charge-discharge characteristics of the all-solid-state cell [6–8]. In contrast, Li3.6Ge0.8S0.2O4 does not form impurities during the sintering process with LiNi1/3Mn1/3Co1/3O2, whether the process is the lowtemperature SPS or high-temperature FS. This result reveals that Li3.6Ge0.8S0.2O4 can be used to fabricate a composite electrode with LiNi1/3Mn1/3Co1/3O2 by FS. Fig. 4a shows the charge-discharge curves for the all-solid-state cell assembled by FS at 800 °C. The electrochemical lithium (de-)intercalation reactions of the LiNi1/3Mn1/3Co1/3O2 cathode are observed in the all-solid-state cell using Li3.6Ge0.8S0.2O4 as electrolyte. The slope 54
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 3. Powder X-ray diffraction patterns of (a) LiNi1/3Mn1/3Co1/3O2, (b) Li3.6Ge0.8S0.2O4, (c) LiNi1/3Mn1/3Co1/3O2/Li3.6Ge0.8S0.2O4 mixture sintered at 900 °C by FS and (d) 600 °C by SPS process.
reaction, although the cell fabricated using SPS delivered higher chargedischarge characteristics than that using FS. Fig. 5 shows the SEM images of the cross-sectional microstructures of the composite electrode/electrolyte sintered at 800 and 900 °C by FS, and 600 °C by SPS. Although the ratio of electrode material/electrolyte sintered was the same in each process, FS yielded a thicker composite electrode layer (ca. 50 μm) than SPS (ca. 25 μm). The calculated porosity of the composite electrode sintered by FS was 39% because only a little uniaxial pressure was applied to the electrode/electrolyte layer during the sintering process. However, the porosity of the sample sintered by SPS could not be calculated correctly because the thickness of the electrode was smaller than that estimated without the pores. A part of the sample might have been peeled off when the sample was removed from the electrode after the SPS experiment to prepare it for SEM. Therefore, the electrode thickness during the battery experiment would be larger. For the composite electrode layer sintered at 800 °C by FS, Fig. 5a shows that LiNi1/3Mn1/3Co1/3O2 particles, with sizes of 1–10 μm, are dispersed with connection between each other, and there are voids between the LiNi1/3Mn1/3Co1/3O2/Li3.6Ge0.8S0.2O4 and Li3.6Ge0.8S0.2O4/Li3.6Ge0.8S0.2O4 phases. Upon increasing the sintering temperature to 900 °C, the domains of the Li3.6Ge0.8S0.2O4 phase become larger and apparent contact between the active material and electrolyte is observed because of the progress in the sintering process (Fig. 5b). In the microstructure of the composite electrode layer sintered by SPS, the LiNi1/3Mn1/3Co1/3O2 particles are dispersed similarly as in FS, and the Li3.6Ge0.8S0.2O4 electrolyte is packed densely. This results in a well-defined interface between the LiNi1/3Mn1/3Co1/3O2/ Li3.6Ge0.8S0.2O4, and Li3.6Ge0.8S0.2O4/Li3.6Ge0.8S0.2O4 phases. The insufficiently connected interfaces of the active material/electrolyte and electrolyte/electrolyte phases in the composite electrode layer sintered by FS could result in a large internal resistance and is one of the causes of the poor charge-discharge performance. Utilization of a LISICONbased material, Li4-2xGe1-xSxO4, that have a γ-Li3PO4-type structure as the solid electrolyte for an all-solid-state battery suppresses the reaction with the active material during the sintering process. Moreover, it permits the fabrication of a cell with a reversible charge-discharge reaction using SPS, which provides dense sintering at a relatively low temperature. The hot-pressing technique has also been reported to
Fig. 4. Charge-discharge curves of the LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 composite cathode/Li3.6Ge0.8S0.2O4 electrolyte/PEO film/lithium foil cell. The layered composite cathode and electrolyte component was sintered at (a) 800 °C and (b) 900 °C by FS process, and (c) 600 °C by SPS process, respectively.
result in dense sintering for oxide ceramics, although it requires a higher temperature for sintering than SPS [21]. Based on our experimental results, Li4-2xGe1-xSxO4 shows high thermal compatibility to LiNi1/3Mn1/3Co1/3O2 during the sintering process. Thus, a high-temperature and high-pressure sintering process such as the hot-pressing method can be applied in the fabrication of an all-solid-state cell with reversible charge-discharge reaction if Li4-2xGe1-xSxO4 is used as the electrolyte.
4. Conclusion The use of the LISICON-based material, Li4-2xGe1-xSxO4, as electrolyte suppressed the reaction with LiNi1/3Mn1/3Co1/3O2 during the cosintering process by FS at temperatures as high as 900 °C and led to the reversible charge-discharge reaction in the solid-state cell. A solid solution was formed for samples with a composition x between 0.1 and 0.4. The Li3.6Ge0.8S0.2O4 sample sintered at 600 °C by SPS exhibited an ionic conductivity of 2 × 10−5 S cm−1 at room temperature, which is 55
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
Fig. 5. Cross sectional SEM images of LiNi1/3Mn1/3Co1/3O2-Li3.6Ge0.8S0.2O4 composite cathode/Li3.6Ge0.8S0.2O4 electrolyte layer sintered at (a) 800 °C and (c) 900 °C by FS, and (e) 600 °C by SPS processes. The expanded images in the composite cathode sintered at (b) 800 °C and (d) 900 °C by FS, and (f) 600 °C by SPS methods.
to thank Ms. Wagou, Mr. Watanabe and Dr. Yamamoto for helping the synthesis, SPS and SEM experiments.
mainly from bulk contribution. This indicates that the resistance derived from grain boundary was decreased by SPS. Although no phase formation was observed in the mixture of LiNi1/3Mn1/3Co1/3O2 and Li3.6Ge0.8S0.2O4 during high-temperature sintering by FS, the assembled cell exhibited limited discharge capacities of 50 and 92 mAh g−1 at 800 and 900 °C, respectively, because of the poor connection between the active material/electrolyte and electrolyte/electrolyte interfaces in the composite layer. In contrast, the cell assembled by SPS had well-connected interfaces and exhibited a larger discharge capacity of 130 mAh g−1 and better initial cycle performance.
References [1] Y. Kato, S. Hori, T. Saito, K. Suzuki, M. Hirayama, A. Mitsui, M. Yonemura, H. Iba, R. Kanno, Nat. Energy 1 (16030) (2016) 1–7. [2] Y. Kato, K. Kawamoto, R. Kanno, M. Hirayama, Electrochemistry 80 (10) (2012) 749–751. [3] A.L. Robinson, J. Janek, MRS Bull. 39 (12) (2014) 1046–1047. [4] A. Sakuda, A. Hayashi, M. Tatsumisago, Sci. Rep. 3 (2261) (2013) 1–5. [5] M. Nose, A. Kato, A. Sakuda, A. Hayashi, M. Tatsumisago, J. Mater. Chem. A 3 (2015) 22061–22065. [6] J. Xie, N. Imanishi, T. Zhang, A. Hirano, Y. Takeda, O. Yamamoto, J. Power Sources 189 (2009) 365–370. [7] Y. Kobayashi, T. Takeuchi, M. Tabuchi, K. Ado, K. Kageyama, J. Power Sources 81–82 (1999) 853–858. [8] K. Park, B.-C. Yu, J.-W. Jung, Y. Li, W. Zhou, H. Gao, S. Son, J.B. Goodenough, Chem. Mater. 28 (21) (2016) 8051–8059. [9] A. Mei, Q.-H. Jiang, Y.-H. Lin, C.-W. Nan, J. Alloys Compd. 486 (2008) 871–875. [10] T. Okumura, T. Takeuchi, H. Kobayashi, J. Ceram. Soc. Jpn. 125 (4) (2017)
Acknowledgements This work was financially supported by the Advanced Low Carbon Technology Research and Development Program of the Japan Science and Technology Agency for Specially Promoted Research for Innovative Next Generation Batteries (JST-ALCA SPRING). The authors would like 56
Solid State Ionics 326 (2018) 52–57
S. Taminato et al.
[17] T. Okumura, S. Taminato, T. Takeuchi, H. Kobayashi, ACS Appl. Energy Mater. (2018) (submitted). [18] F. Izumi, K. Momma, Solid State Phenom. 130 (2007) 15–20. [19] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751–767. [20] N. Yabuuchi, Y. Makimura, T. Ohzuku, J. Electrochem. Soc. 154 (4) (2007) A314–A321. [21] S.B. Li, C.B. Wang, D.Q. Zhou, H.X. Liu, L. Li, Q. Shen, L.M. Zhang, Ceram. Int. 44 (1) (2018) 550–555.
276–280. [11] T. Okumura, T. Takeuchi, H. Kobayashi, Solid State Ionics 188 (2016) 248–252. [12] M.A.K.L. Dissanayake, R.P. Gunawardane, A.R. West, G.K.R. Senadeera, P.W.S.K. Bandaranayake, M.A. Careem, Solid State Ionics 62 (3–4) (1993) 217–223. [13] H.Y.-P. Hong, Mater. Res. Bull. 13 (2) (1978) 117–124. [14] J.G. Kamphorst, E.E. Hellstrom, Solid State Ionics 1 (3–4) (1980) 187–197. [15] J. Kuwano, A.R. West, Mater. Res. Bull. 15 (11) (1980) 1661–1667. [16] A.R. Rodger, J. Kuwano, A.R. West, Solid State Ionics 15 (3) (1985) 185–198.
57