TiO₂ Anode Material for All-Solid-State Battery Using NASICON Li_1.5Al_0.5Ge_1.5(PO₄)₃ as Lithium Ion Conductor

Abstract

We have been developing sintered multilayer oxide-based all-solid-state batteries. Anode active material rutile-type TiO₂ was not reacted with amorphous Na superionic conductor (NASICON)-type solid electrolyte Li_1.5Al_0.5Ge_1.5(PO₄)₃ (LAGP) even after sintered at 600 °C in a nitrogen atmosphere from the XRD patterns. The charge/discharge behavior of the electrochemical measuring cell (when using a non-aqueous electrolyte) was not different from that of rutile-type TiO₂. However, anatase-type TiO₂ charge/discharge behavior changed after sintering process. Additionally, in the result of the input/output characteristics using multilayer oxide-based all-solid-state battery, rutile-type TiO₂ as anode material was 3 times higher discharge capacity than anatase-type TiO₂ at current value 25.6 µA mm⁻². Finally, we successfully measured the Raman spectroscopy of all-solid-battery and rutile-type TiO₂ Raman shift peaks were reversibility during charge/discharge. Based on these findings, we conclude that rutile-type TiO₂ maintained a strong crystalline structure and high Li diffusivity even when sintered with amorphous LAGP. It suggested that rutile-type TiO₂ is suitable as anode material for oxide-based all-solid-state batteries requiring the sintering process.

1. Introduction

Lithium-ion (Li⁺) secondary batteries utilizing nonaqueous electrolytes exhibit high energy densities, rendering them the preferred power storage solution for vehicles, mobile devices, and energy storage systems. However, as the energy density of these batteries increases, their safe management becomes more challenging. In particular, the increase in reports of explosions involving such devices attests this issue. Thus, there has been growing interest in all-solid-state batteries because of the excellent manufacturing safety owing to their non-flammable solid electrolytes. Sulfide-based and oxide-based solid electrolytes are the main areas of research. Sulfide-based solid electrolytes have exhibited plasticity and Li⁺ conductivity comparable to that of conventional non-aqueous electrolytes; however, they are not water-resistant and can react with air moisture to generate hydrogen sulfide. Therefore, it is necessary to implement more stringent moisture management protocols, exceeding even the dew point of −60 °C. Whereas, oxide-based solid electrolytes are easy to treat and are safer in the air atmosphere. However, issues arise due to low ion conductivity and high interface resistance between solid electrolytes and active materials. Furthermore, the sintering process causes the problem of the side reaction with each materials.

Our previous research has focused on oxide-based solid electrolytes, and we have been developing sintered multilayer oxide-based all-solid-state batteries using a Na superionic conductor (NASICON)-type solid electrolyte Li_1.5Al_0.5Ge_1.5(PO₄)₃ (LAGP).¹ These batteries utilize Li₂CoP₂O₇ (LCPO) as the positive electrode active material that can operate at a high voltage of 5 V versus Li⁺/Li.² LCPO cannot be used in organic liquid-based batteries because it causes oxidative decomposition of the electrolyte owing to its high potential. Therefore, LCPO could be effectively utilized in an all-solid-state battery containing LAGP, as this Li⁺ conductor exhibits oxidative decomposition resistance at high voltage. Accordingly, our all-solid-state battery using LCPO and LAGP can perform both charge/discharge operations.³ In contrast, the anode material used in our all-solid-state battery was anatase-type TiO₂, which operates at a voltage between 1.7 and 1.9 V versus Li⁺/Li. However, anatase-type TiO₂ may react during the sintering process with LAGP because the anatase-type TiO₂ crystal structure is unstable at over 800 °C. In general, anatase-type TiO₂ does not have the most stable crystalline structure and undergoes a phase change to rutile-type TiO₂ at 800–1000 °C. Furthermore, Li_1.4Al_0.4Ti_1.6(PO₄)₃ (LATP) may be produced by the elemental substitution of Ti in Ge and TiO₂ in LAGP. The resulting product of LATP is an oxide-based solid electrolyte that operates as an active material at 2.5 V versus Li⁺/Li.⁴ The cell design would be significantly difficult if both anatase-type TiO₂ and LATP act as active materials for the negative electrode in an all-solid-state battery. In the long-term charge/discharge cycle test, a shift in the charge/discharge depth between the positive and negative electrodes can be expected. In addition, it is difficult to control the unexpected charge/discharge behavior of LATP generated by the sintering process. Therefore, the active material for the sintered all-solid-state battery must have a stable structure without reacting with LAGP.

The rutile-type TiO₂ crystal structure is thermodynamically stronger than that of anatase-type TiO₂. The rutile-type TiO₂ is expected to maintain its crystalline structure even when sintering with LAGP. In this study, we prepared the sintered multilayer oxide-based all-solid-state batteries using rutile-type TiO₂ as anode material and compared their charge/discharge performances and those using anatase-type TiO₂. Furthermore, the charge/discharge behavior of rutile-type TiO₂ was investigated by a Raman mapping analysis of the negative electrode layer of an all-solid-state battery. In particular, we studied the relationship between the charge/discharge properties and the capacity.

2. Experimental

The anode materials used in this study were anatase-type TiO₂ (Sakai Chemical Industry Co., Ltd.) and crystalline rutile-type TiO₂ (Sakai Chemical Industry Co., Ltd.) (Figs. S1a and S1b).⁵ The cathode material LCPO was synthesized via a solid phase reaction.² The solid electrolyte was amorphous LAGP (synthesized by the melt quenching method), with a particle size range of 0.2–1 µm.¹ The LAGP amorphous phase changed to Li_1.5Al_0.5Ge_1.5(PO₄)₃ crystalline phase at 600 °C in the sintering process. The crystalline LAGP exhibited ion conductivity 2.0 × 10⁻⁵ S cm⁻¹ at 25 °C, activation energy for Li⁺ transport has 33.9 kJ mol⁻¹.

The crystal structure was determined via X-ray Diffraction (XRD) with a CuKα tube sphere as the source (XRD, SmartLab, Rigaku Corp.). The crystal structure and space group of the active material and solid electrolyte were identified as follows: anatase-type TiO₂ possesses a tetragonal crystal structure (ICSD No. 154604, space group I41/amd), and rutile-type TiO₂ exhibits a tetragonal crystal structure (ICSD No. 202241, space group P42/mnm). LCPO was found to match the monoclinic peak pattern (ICSD No. 261899, space group P21/a), and LAGP matched the peak pattern of the trident crystal (ICSD No. 69763, space group R-3c) of LiGe₂(PO₄)₃ of NASICON. The particle size was observed via field-emission scanning electron microscopy (FE-SEM; JSM-IT700HR, JEOL Ltd.).

Electrode pellets for cells using a non-aqueous electrolyte were prepared using the active materials, Acetylene Black (AB HS-100, Denka) and polyethylene terephthalate (PTFE, Aldrich), in a weight ratio of 25 : 60 : 15 wt%. Electrode pellets were pressed into pellets of 8 mm diameter using a press equipment at a pressure of 15 kN. The electrode thickness was 500 µm, and the electrode density was 1.0 g cm⁻². The electrochemical measuring cell (when using a non-aqueous electrolyte) was an in-house manufactured equipment capable of pressurizing 0.7 kgf. The organic solvent was EC/EMC (30 : 70 vol%) (Tomiyama Pure Chemical Industries, Ltd.), in which 1 mol L⁻¹ LiPF₆ was dissolved as a solid electrolyte. A 20-µm-thick copper foil (Furukawa Electric Co., Ltd.) was used as the collecting foil on the anode side. The anode was a 750-µm-thick lithium (Li) metal foil (Honjo Metal Co., Ltd.) crimped on copper foil. The charging/discharging device used was SD8 (Hokuto Denko Co., Ltd.). In a cell using a non-aqueous electrolyte, the direction in which Li is inserted into the active material is the charge process, and the direction in which Li is desorbed is the discharge process. The charging step was executed using a constant current (CC) of 3.0 V versus Li⁺/Li endpoint, and the discharge step was conducted at 1.25 V versus Li⁺/Li endpoint at a temperature of 25 °C with a current density of 335 mA h g⁻¹ as 1 C. When measuring anatase-type TiO₂, a capacity termination condition of 100 mA h g⁻¹ was also incorporated into the discharge termination. The thermal stability of TiO₂, which was evaluated by mixing with a solid electrolyte and sintered, was set to 0.1 C for 10 cycles. The impedance measurement was performed using Solartron 1287A as the potentiostat and Solartron 1255B as the frequency response analyzer (Solartron Analytical). The impedance measurement was performed in the frequency range of 1 MHz–0.1 Hz with an amplitude of 5 mV (Fig. S3).

The all-solid-state battery in this study has a layered construction similar to that of a multilayer ceramic capacitor.^6,7 First, to obtain electrolyte green sheet, add 30 wt% of binder PVDF to the LAGP amorphous powder 40 wt% of anhydrous alcohol as ethanol as a solvent to the LAGP amorphous powder, and mixed the raw materials in paste form. In order to uniformly mixed the raw materials of electrolyte layer materials, the paste were mixed in a ball mill for 20 h. After defoaming in a vacuum, the paste was coated on PET film by the doctor blade method to obtain one layer of electrolyte layer material formed into a sheet. In order to adjust the electrolyte layer sheet to the 20 µm thickness, one coated sheet obtained in one coating was stacked for a specified number of layers. The coating process is repeated, and once the electrolyte layer material has been formed into a sheet of the desired thickness, the sheet is press-bonded, and the sheet after the bonding was cut to the specified flat size to complete the electrolyte layer sheet. Next, to prepare the active material pastes, LCPO was used as the active material for the positive electrode and TiO₂ for the negative electrode. The active material, amorphous LAGP powder, a binder as acrylic resin, VGCF (vapor-grown carbon fiber, Resonac) as carbon-based conductivity aid, and ethanol as solvent are mixed to form the active material pastes. These pastes were adjusted to 50 µm thickness and the same amount of active material for cathode layer and anode layer on one side of the green sheet of solid electrolyte, and coated by screen printing. After coating, the paste was dried in a dryer at a temperature of 100 °C for 30 minutes. These amounts of active materials capacity contained cathode layer and anode layer used the same ratio of LCPO and TiO₂. The green sheets of LAGP with printed positive electrode alternately stacked on top of green sheets of LAGP with printed negative electrode. After then, the stacked green sheets were cut to chip size which is cuboid. In the sintering process, after removing organic components from the green chips, the all-solid-state battery was sintered at 600 °C in a nitrogen atmosphere. After the sintering process, solid state battery’s solid electrolyte thickness was between 6 and 9 µm, electrode thickness was between 15 and 25 µm, and the electrode density of the positive and negative electrode was 0.3 g cm⁻².

In this experiment, the charge/discharge capacity of the all-solid-state battery was limited by the capacity of the cathode active material, LCPO. During the charging steps, Li moved from the cathode side to anode side, and the opposite reaction occurred during discharge. The cells were fabricated by adjusting the amount of cathode active material capacity, such that the charge state of the anode material was 20 %. This charge depth capacity corresponds to 120 µA h (70 mA h g⁻¹). The charge/discharge current value was 1.6 µA mm⁻² between 1 and 10 cycles; the charge step was performed at a Constant Current (CC) of 3.6 V stop, and the discharge step was charged and discharged at the CC 0 V stop. Thereafter, the charge/discharge current value was increased to 3.2, 6.4, 12.8, and 25.6 µA mm⁻² every 10 cycles, and the current value was set to 1.6 µA mm⁻² between 51 and 55 cycles to investigate the charge/discharge rate characteristics.

The cross-sectional processing layer of the all-solid-state batteries was prepared using a cross-sectional polisher device (IB-19510-CP, JEOL Ltd.). For spectroscopic analysis, a microscopic laser Raman spectrometer (LabRAM HR Evolution VIS-NIR, HORIBA, Ltd.) was used to measure the Raman spectrum to observe the rutile-type TiO₂ contained in the anode layer of the all-solid-state battery. Raman measurements were performed before charging and discharging at room temperature. Charging was subsequently performed at a CCCV of 3 V and a current density of 16 µA mm⁻², and charging was stopped at a charge capacity of 130 µA h. Raman measurements were performed during a 40-min pause after charging. Next, discharge was performed at a CC 0 V termination with a current density of 1.6 µA mm⁻². Raman measurements were performed during a 40-minute pause after discharge. The Raman spectra were collected in the range 320–540 cm⁻¹ by laser with a wavelength of 457 nm and acquisition time was 1 second at each measurement point.

3. Results and Discussion

First, the charge/discharge behavior was obtained by investigating the rate characteristics of a non-aqueous electrolyte cell using Li metal as the anode for rutile-type TiO₂ (Fig. 1a). It has been reported that the rutile-type TiO₂ forms an inert Li phase (irregular monoclinic α-NaFeO₂ type Li_xTiO₂) near 1.1 V versus Li⁺/Li, which causes irreversible capacity. Therefore, the discharge cutoff voltage was limited to 1.25 V versus Li⁺/Li.⁸ When a charge/discharge test with a limited capacity was performed, Li storage ended near 1.25 V versus Li⁺/Li before formation into an irreversible phase occurs by regulating the amount of Li insertion, changing the Li desorption reaction.⁹

Figure 1.

Charge/discharge curves at a current density of 33.5 mA g⁻¹ (0.1 C) between 0.1 and 2 C for the (a) Rutile-type TiO₂ electrode, (b) Anatase-type TiO₂ electrode, and (c) Rate capabilities of rutile- and anatase-type TiO₂.

Figure 1b shows the charge/discharge behavior of the anatase-type TiO₂. Figure 1c shows the transition of the discharge capacity at different charge/discharge rates. The anatase-type TiO₂ significantly decreased the charging capacity by reaching a terminal voltage of 1.25 V versus Li⁺/Li via increasing the discharge rate. In contrast, the rutile-type TiO₂ tended to exhibit an excellent rate response.

Subsequently, the thermal stability of TiO₂ was evaluated when it was mixed with a solid electrolyte and sintered. After mixing the TiO₂ anode active material and the amorphous solid electrolyte LAGP at a weight ratio of 1 : 1, a coating electrode was prepared using a powder sintered in a nitrogen atmosphere at 600 °C, and a charge/discharge test was performed on the non-aqueous electrolyte cell. Figure 2a shows the charge/discharge behavior of the cells using anatase-type TiO₂. The cell of anatase-type TiO₂ showed a plateau charge/discharge curve between 1.7 and 1.9 V for Li⁺/Li. The charge/discharge behavior of anatase-type TiO₂ is formed by Li insertion, while the charge/discharge drops from the open-circuit voltage to 1.7 V versus Li⁺/Li. Subsequently, a two-phase coexistence state is seen between TiO₂ and tetragonal LiTiO₂, and a clear flat plateau is observed.

Figure 2.

Charge/discharge curves at a current density of 33.5 mA g⁻¹ (0.1 C) during 10 cycles for the (a) Anatase-type TiO₂ electrode, (b) Rutile-type TiO₂ electrode, (c) Anatase-type TiO₂ electrode sintered with amorphous LAGP, and (d) Rutile-type TiO₂ electrode sintered with amorphous LAGP.

The discharge cutoff voltage of the cell using rutile-type TiO₂ was 1.25 V versus Li⁺/Li (Fig. 2b). The rutile-type TiO₂ exhibits smooth shoulder curves between 1.25 and 2 V versus Li⁺/Li. In the charge/discharge behavior of rutile-type TiO₂, the Li diffusion path exists only in the c-axis direction, TiO₂ exhibits low electron conductivity, and the charge/discharge curve forms a gentle shoulder.

Regarding the sintered anode materials, the anatase-type TiO₂ sintered with amorphous LAGP exhibited a shoulder in the charge/discharge curves between 2 and 2.5 V versus Li⁺/Li (Fig. 2c). The rutile-type TiO₂, even when sintered with amorphous LAGP, has a little change in the charge/discharge behavior at 2.5 V versus Li⁺/Li was observed (Fig. 2d). It suggest that the grain boundary between anode materials and LAGP electrolyte have changed after sintering process. These results shows the rutile-type TiO₂ has more thermal stability compere with anatase-type TiO₂.

The XRD measurement results of the powder exhibit no impurity phases, which seem to be LATP,³ observed before and after sintering of the negative electrode active material (Figs. S2a, S2b, S2c, S2d, S2e, S2f, S2g, and S2h). The rutile-type TiO₂ maintains a strong crystalline structure, and no change in the charge/discharge behavior is exhibited. Regarding the charge/discharge efficiency of rutile-type TiO₂, after co-sintering with amorphous LAGP, the first charge/discharge efficiency exhibited an irreversible capacity. The first charge/discharge efficiency was at approximately 60 %, and the second charge/discharge efficiency was more than 90 %. The previous study suggest that rutile-type TiO₂ irreversible capacity have been attributed to Li inert phase (irregular monoclinic α-NaFeO₂ type Li_xTiO₂).⁸ Therefore, it assumed that Li was inserted into the rutile-type TiO₂ by co-sintering with the amorphous LAGP solid electrolyte, and an Li inert phase was generated between the grain boundaries of TiO₂ and LAGP.

The all-solid-state battery used anatase- and rutile-type TiO₂ as the anode materials. Both all-solid-state batteries used Li₂CoP₂O₇ as the cathode material. The behavior of the charge/discharge test at a current value of 1.6 µA mm⁻² is shown in Figs. 3a and 3b. The charging step was CC 3.6 V cutoff, discharging step was CC 0 V cutoff, and all-solid-state batteries using rutile-type TiO₂ were not charged any further beyond the 130 µA h charging capacity of first and second cycles.

Figure 3.

Solid-state battery’s charge/discharge curves at a current density of 1.6 µA mm⁻² during three cycles for the (a) Anatase-type TiO₂ electrode and (b) Rutile-type TiO₂ electrode as anode material. Solid-state battery’s discharge curves between 1.6 µA mm⁻² and 25.6 µA mm⁻² for the (c) Anatase-type TiO₂ electrode and (d) Rutile-type TiO₂ electrode as anode material. (e) Rate capabilities of the solid-state batteries between 1.6 µA mm⁻² and 25.6 µA mm⁻².

In the first charge/discharge behavior in all-solid-state batteries using anatase-type TiO₂, a shoulder curve was observed around 2.5 V. This behavior is the same as the charge/discharge results of the non-aqueous electrolyte cell. Furthermore, no 2.5 V shoulder curves were observed for all-solid-state batteries using rutile-type TiO₂. For the charge/discharge efficiency, a reversible capacity of approximately 80 % was observed only for the first time, as in the non-aqueous electrolyte cell, and subsequently the charge/discharge efficiency of 90 % or more was observed after the second cycle.

Subsequently, to investigate the input/output characteristics, the result of charge/discharge current value increased from 1.6 µA mm⁻² to 25.6 µA mm⁻² every 10 cycles, as shown in Figs. 3c, 3d, and 3e. The rutile-type TiO₂ showed a charge/discharge behavior superior to that of anatase-type TiO₂. In particular, at a current value of 25.6 µA mm⁻², the discharge capacity difference of rutile-type TiO₂ was three times higher than that of anatase-type TiO₂, and it was confirmed that the rutile-type TiO₂ anode material exhibits excellent rate characteristics in all-solid-state batteries. Furthermore, the irreversible capacity of solid-state battery has improved compared with the electrochemical measuring cell. It suggest that the irreversible capacity involved organic solvent in the test of the electrochemical measuring cell.

The evaluation results of the all-solid-state battery of rutile-type TiO₂ revealed that it had better rate characteristics, than anatase-type TiO₂, while the average discharge voltage was 2.5 V. It would suggest that this is due to the fact that an inert Li phase (irregular monoclinic α-NaFeO₂ type Li_xTiO₂) is generated, and the operating area of the charge/discharge changes. To confirm of this consideration, it need to analyze the atomic arrangement of the grain boundaries between LAGP and the crystalline phase of the anode active material.

The Raman mapping results obtained for the all-solid-state battery using rutile-type TiO₂ as the negative electrode, normalized by the Raman peak of 430 cm⁻¹ of rutile-type TiO₂ before and after charging and discharging, are shown in Fig. 4 (Figs. 4a, 4b, 4c, 4d, and 4e). Two cycles of charging and discharging were repeated under these charging and discharging conditions to obtain the Raman shift of the rutile-type TiO₂ all-solid-state battery (Figs. 4f and 4g). Comparing the Raman shifts before and after charging and discharging, which exhibited the strongest peak intensity on the measurement plane, a decrease in the peak intensity and a shift to the low wavenumber side were observed upon charging, and a return to the original position upon discharging was captured. The decrease in Raman shift intensity due to Li insertion into rutile-type TiO₂ was previously demonstrated by Christensen et al. through a chemical approach,⁸ and we considered that this result have demonstrated in an all-solid-state battery in this study. The Raman peak may have shifted due to stress distortion caused by expansion and contraction associated with charging and discharging. However, these charge/discharge behaviors during the rate characteristic study and the Raman measurement were slightly different, probably due to the cross-sectional processing of the all-solid-state battery. Therefore, the impact of processing for charge/discharge behavior should be considered concurrently.

Figure 4.

Solid state battery’s Raman mapping results and Raman shift peak (430 cm⁻¹) for the (a) as-prepared battery, (b), (c) after first charge/discharge, and (d), (e) after second charge and discharge. (f) Raman shift peaks indicate the Raman wavenumber at each charge/discharge step. (g) charge/discharge curves during Raman measurements.

Finally, the phenomenon of charge/discharge voltage in all-solid-state batteries of anatase- and rutile-type TiO₂ was investigated in a half-cell using Li metal. A green sheet consisting of only the anode layer used in the all-solid-state battery was prepared; further, it was sintered under the same sintering conditions as those used in the all-solid-state battery to obtain a sintered anode sheet. After depositing gold as a current-collecting layer on one side, Li metal was placed through a polymer electrolyte containing LiTFSI on the opposite side from the gold current-collecting layer and sealed in a laminate to make a half cell. C (3.35 mA g⁻¹) at 105 °C (Fig. S4). The charge/discharge behavior in both cells was confirmed, suggesting that the anode operation was reflected in the all-solid-state battery. Pertaining to the cause of the change in operating voltage, certain change is considered to occur at the interface between the solid electrolyte LAGP and the negative electrode active material. However, since XRD and Raman measurements did not confirm the presence of a different phase, the change in behavior is considered to be caused by an interface layer of the order of a few nanometers. The high Li diffusivity of rutile-type TiO₂ can be achieved in all-solid-state batteries. Further detailed investigation will reveal the cause of the change in the voltage behavior and the change in the Li diffusivity of rutile-type TiO₂ in all-solid-state batteries.

4. Conclusions

It was confirmed that rutile-type TiO₂ as an anode active material exhibited better input/output characteristics than anatase-type TiO₂, in both electrolyte-based cells and all-solid-state batteries. In all-solid-state batteries, the charge/discharge voltage curves of both anode active materials displayed changes; however, the Li diffusivity of rutile-type TiO₂ remained higher than that of anatase-type TiO₂. This is an important result, showing that Li diffusivity in the bulk of the anode active material can be maintained even after co-firing with a solid electrolyte. We have succeeded in visualizing the reaction distribution of Li deinsertion in rutile-type TiO₂ to analyze a cross section of an all-solid-state oxide battery by using Raman spectroscopy. This technique enabled visually understanding the presence or absence of the reaction distribution of active materials in the anode layer of an all-solid-state battery. Further detailed investigation of the solid-state interface will advance our understanding of the reaction behavior during charging and discharging. It should be necessary to investigate the grain boundaries between LAGP and the crystalline phase of the anode active material by analyzing the elemental diffusion at the solid electrolyte-anode active material interface by TEM-EELS and the atomic arrangement by HAADF STEM and ABF STEM for the solid interfaces inside all solid-state batteries on next works.

Data Availability Statement

The data that support the findings of this study are openly available under the terms of the designated Creative Commons License in J-STAGE Data at https://doi.org/10.50892/data.electrochemistry.22884665.

• 1) We have been developing sintered multilayer oxide-based all-solid-state batteries. Anode active material rutile-type TiO₂ was not reacted with amorphous Na superionic conductor (NASICON)-type solid electrolyte Li_1.5Al_0.5Ge_1.5(PO₄)₃ (LAGP) even after sintered at 600 °C in a nitrogen atmosphere from the XRD patterns. The charge/discharge behavior of the electrochemical measuring cell (when using a non-aqueous electrolyte) was not different from that of rutile-type TiO₂. However, anatase-type TiO₂ charge/discharge behavior changed after sintering process. Additionally, in the result of the input/output characteristics using multilayer oxide-based all-solid-state battery, rutile-type TiO₂ as anode material was 3 times higher discharge capacity than anatase-type TiO₂ at current value 25.6 µA mm⁻². Finally, we successfully measured the Raman spectroscopy of all-solid-battery and rutile-type TiO₂ Raman shift peaks were reversibility during charge/discharge. Based on these findings, we conclude that rutile-type TiO₂ maintained a strong crystalline structure and high Li diffusivity even when sintered with amorphous LAGP. It suggested that rutile-type TiO₂ is suitable as anode material for oxide-based all-solid-state batteries requiring the sintering process.

CRediT Authorship Contribution Statement

Yoichiro Kawano: Data curation (Lead), Formal analysis (Lead), Investigation (Lead), Writing – original draft (Lead), Writing – review & editing (Lead)

Akihiko Kato: Project administration (Equal)

Hiroyuki Usui: Project administration (Equal)

Yasuhiro Domi: Methodology (Equal)

Hiroki Sakaguchi: Conceptualization (Lead)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Footnotes

A part of this paper has been presented in the 63rd battery Symposium in Japan in 2022 (Presentation #2E05).

Y. Kawano, A. Kato, H. Usui, and Y. Domi: ECSJ Active Members

H. Sakaguchi: ECSJ Fellow

References

• 1)  J. K. Feng, L. Lu, and M. O. Lai, J. Alloys Compd., 501, 255 (2010).
• 2)  H. Kim, S. Lee, Y.-U. Park, H. Kim, J. Kim, S. Jeon, and K. Kang, Chem. Mater., 23, 3930 (2011).
• 3)  C. Yamamoto, S. Doi, M. Nakanishi, M. Kobayashi, M. Inagaki, Y. Goto, J. Kato, K. Yamamoto, A. Kato, and Y. Kataoka, Abstracts of PRiME2020, A05-0911 (2020).
• 4)  S. Patoux and C. Masquelier, Chem. Mater., 14, 5057 (2002).
• 5)  H. Usui, Y. Domi, S. Ohnishi, N. Takamori, S. Izaki, N. Morimoto, K. Yamanaka, K. Kobayashi, and H. Sakaguchi, ACS Mater. Lett., 3, 372 (2021).
• 6)  G. Okuma, N. Saito, K. Mizuno, Y. Iwazaki, H. Kishi, A. Takeuchi, M. Uesugi, K. Uesugi, and F. Wakai, Acta Mater., 206, 116605 (2021).
• 7)  H. Masuda, N. Ishida, Y. Ogata, D. Ito, and D. Fujita, J. Power Sources, 400, 527 (2018).
• 8)  C. K. Christensen, M. A. H. Mamakhel, A. R. Balakrishna, B. B. Iversen, Y. M. Chiang, and D. B. Ravnsbæk, Nanoscale, 11, 12347 (2019).
• 9)  H. Usui, Y. Domi, T. H. Nguyen, S. Izaki, K. Nishikawa, T. Tanaka, and H. Sakaguchi, Electrochemistry, 90, 037002 (2022).