MATERIALS TRANSACTIONS
Online ISSN : 1347-5320
Print ISSN : 1345-9678
ISSN-L : 1345-9678
Materials Chemistry
Preparation and Electrochemical Properties of LiCoO2 Electrode Layer by Molten Salts on Mechanical Machined Li0.29La0.57TiO3 Solid Electrolyte
Hijiri OikawaTakanori YamamotoYoshinori Arachi
Author information
JOURNAL FREE ACCESS FULL-TEXT HTML

2022 Volume 63 Issue 8 Pages 1159-1163

Details
Abstract

An effect of surface area and microstructure of Li0.29La0.57TiO3 (LLTO) solid electrolyte on an electrochemical performance of LiCoO2 (LCO) as positive electrode in an all-solid-state Li secondary battery was examined. The surface of LLTO sintered body was machined mechanically by using a pico-second laser and obtained in various forms of LLTO. Then, LCO on the LLTO was prepared from molten salts as a starting material. The microstructural observation confirmed that LCO was formed on the surface of a laser-machined LLTO. A discharge capacity was enlarged with the amount of LCO loading. These results suggest that the LCO formed on LLTO by molten salts was active electrochemically regardless of various shapes of microstructures of the solid electrolyte.

 

This Paper was Originally Published in Japanese in J. Jpn. Soc. Powder Powder Metallurgy 69 (2022) 104–107.

1. Introduction

In recent years the research and development on all-solid-state Li secondary batteries (ASSBs) has been actively promoted in which organic electrolyte solutions are replaced by solid electrolytes. The ASSBs using sulfide-based electrolytes are about to come into practical applications as power sources for electric vehicles,1) whereas those using oxide-based ones are limited for small electronic devices despite their excellent safety. The difference is due to at least two decisive factors. Firstly, oxide-based electrolytes have an ionic conductivity in the 10−4∼10−5 Scm−1 range at room temperature that is two orders of magnitude lower than those of sulfide-based ones. Secondary, there is a much higher interfacial resistance between electrodes and solid electrolytes in oxide-based ASSBs than in sulfide-based ones. The latter allow bonding between them by pressing at room temperature. In contrast, the oxide-based ASSBs require heating at high temperatures to form an electrochemical active interface.

Therefore, it is crucial to develop a process to form the interface with depressed resistance by repressing an inter-diffusion. Sputtering or vapor deposition methods have been employed to produce thin film positive electrodes with a several µm in thickness on oxide-based electrolyte substrates.24) On the other hand, a sintering at low temperature has attempted by addition of Li3BO3 with melting point of 973 K, followed by screen-printing LiCoO2 (LCO) slurry on Nb doped Li7La3Zr2O12 solid electrolyte with a cubic garnet-type structure.5) The resultant LCO with 10 µm in thickness exhibits a low interfacial resistance and good electrochemical performance as positive electrode in ASSBs.

However, towards a wide spread and extend applications of the oxide-based ASSBs, it is essential to enhance the battery capacity and energy density by increasing the amount of active material. Previously, formations of LCO by molten salt at lower temperatures and the electrochemical properties in conventional liquid lithium secondary batteries (LiBs) have been reported.68) The 0.3 µm thickness of LCO film is formed on Au foil by the oxidation reaction of Co metal in the molten carbonate. The LCO produced by LiNO3 and LiCl eutectic molten salt shows a reversible capacity of 167 mAh g−1 in the voltage range from 2.5 to 4.4 V. The LCO is generated on Pt foil using a mixed flux of LiNO3 and LiOH. In all these reports an excess Li salt is used as a flux to grow the LCO crystal and the electrochemical properties of these LCO is examined in LiBs.

On the other hand, we have paid attention to fluidity of molten salts as a starting material during heating to form an electrochemical active interface between oxides, focusing on ASSBs consisting of active material of positive electrode and solid electrolyte without any conductive assistants. Our proposed process involves placing mixed lithium and cobalt salts on Li0.29La0.57TiO3 (LLTO) solid electrolyte and heating at 973 K for 1 h, thereby creating the LCO with high electrochemical activity.9)

In the present study the process was applied to on various LLTO solid electrolytes with different surface shapes, and evaluated the electrochemical activity of the LCO produced from the molten salts in an all-solid-state Li secondary battery. LLTO sintered body was machined mechanically to various shapes by irradiating with laser light to change the surface structure of the LLTO. Then, LCO as the positive electrode was generated on the LLTO solid electrolyte through the process. Here, we present an effect of various surfaces of the LLTO solid electrolyte on the electrochemical activity of the LCO formed from molten salts in ASSBs.

2. Experimental Procedure

2.1 Process and materials

The surface of Li0.29La0.57TiO3 sintered body10) (t0.5 mm × ϕ10 mm, TOHO TITANIUM Co., Ltd.) was machined using a picosecond laser with a wavelength of 532 nm. An optical microscope (OM) was used to observe the microstructures of the mechanically-machined samples. The machined LLTO samples were heated at 873 K for 1.8 ks at heating rate 1/6 Ks−1 in ambient atmosphere. LCO was synthesized according to our previous paper.9) The starting materials were LiNO3 (>99.9%, Fujifilm Wako Chem.), LiCl (>99.9%, Fujifilm Wako Chem.), and Co(NO3)2·6H2O (>99.9%, Fujifilm Wako Chem.). A eutectic mixture of LiNO3:LiCl in a molar ratio of 0.88:0.12 was used as the Li-source. A stoichiometric mixture of the Li and Co salts was placed on the machined LLTO sample and then it was sintered at 973 K for 3.6 ks to generate LCO on the LLTO solid electrolytes.

2.2 Characterizations

X-ray diffraction (XRD) measurements were performed using CuKα radiation to confirm the crystal phases (XRD-6100, Shimadzu Co. Ltd.). FE-SEM (JSM-6700, JEOL) observations of the surfaces of the machined LLTO samples and EDS (energy dispersive spectroscopy) for elemental mappings were conducted. The electrochemical properties of the LCO electrodes on various machined LLTO solid electrolytes were examined with an all-solid-state cell tester. Au sputtering as a current collector was carried out on the LCO side using a quick coater (Sanyu Electronics Co., Ltd., SC-701MkII). An In–Li alloy was used as the anode electrode. The electrode potential of the alloy is 0.62 V vs. Li+/Li. Between LLTO and anode electrode Li3.25Ge0.25P0.75S4 (LGPS) powder was inserted to prevent the reduction of LLTO by Li and to assure the mobility of Li ions. The LGPS was prepared according to the previous report by Kanno et al.11) Charge-discharge testing was conducted galvanostatically at 323 K, while applying various current densities from 0.01 C to 0.1 C-rate, where a current of 137 mA g−1 was defined as 1 C-rate, in voltage range from 2.0 to 3.6 V corresponding to 2.6 to 4.2 V vs. Li+/Li. The weight of LCO was calculated by subtracting the weight before immersing LCO from that after filling LLTO with LCO. The experimental error of specific capacity was around 1–2 mA hg−1.

3. Results and Discussion

3.1 Machining of LLTO and LCO formation

The surface of LLTO sintered body was machined mechanically under various conditions using a picosecond laser with a wavelength of 532 nm (Table 1). Two types of shapes, lattice and stripe were adopted in different dimensions, leading for LLTO sintered body to have different surface areas. The surfaces of the machined LLTO samples were observed by optical microscope. The optical photographs of LLT-mm1 and LLT-mm2 are shown in Fig. 1. The surfaces of the samples were processed to desired shapes, whereas the bottoms of all machined samples were not in flat, but pointed shapes, as described later. XRD measurements were examined to observe some changes in crystal structure caused by the mechanical machining and annealing at 873 K for 1.8 ks (Fig. 2). The appearance photographs of the samples were inserted into the figure. The color of only the processed part of the LLTO sample changed from white to black. However, annealing at 873 K for 1.8 ks in Ar flow made the sample return to the original white color. In the LLTO sample any impurity phases were not observed. From the calculated tetragonal lattice parameters of LLTO, the a-axis decreased and c-axis increased slightly by the mechanical machining and then they returned to the original lattice parameters (Table 2). The laser irradiation may have caused Ti4+ in LLTO to reduce to Ti3+ and lead the lattice expansion of the LLTO. Since the lattice parameters returned to the original ones by the annealing, the starting materials of LiCoO2 (LCO) was filled on the laser-machined LLTO samples and followed by heating at 973 K for 3.6 ks. Figure 3 shows XRD patterns for LiCoO2 obtained on different LLTO samples: (a) before machining, (b) after machining (LLT-mm3). All the peaks were indexed to the trigonal LiCoO2 at hexagonal setting except for those of the sample holder and LLTO (LLT-mm3). The lattice parameters were calculated as a = b = 2.83621(11) Å, c = 14.0511(71) Å. FE-SEM microstructure images of the cross section and surface of LLTO immersed with LCO are shown in Figs. 4(a) and (c), and the EDS elemental mappings of Co-K, La-L and Ti-K in Figs. 4(b-1), (b-2), (b-3). The Co was distributed widely on the surface of the laser-machined LLTO in depth of 80 to 100 µm, in addition to the top of the LLTO. The LCO was generated in these parts. The amount of LCO loading differed depending on the mechanical processing shape of LLTO, as described later in details. It should be noticed that the LCO did not fill the grooved LLTO, but the surface of the LLTO. The shape of the laser-machined LLTO mostly remained unchanged.

Table 1 Summary of mechanical machined-LLTO sintered body.
Fig. 1

Optical photograph of (a) LLT-mm1 and (b) LLT-mm2.

Fig. 2

XRD patterns of LLTO substrates: (a) bare, (b) machined mechanically (LLT-mm3) and (c) annealed at 873 K for 1.8 ks in Ar flow (LLT-mm3).

Table 2 Summary of calculated tetragonal LLTO lattice parameters: (a) bare, (b) machined mechanically (LLT-mm3) and (c) annealed at 873 K for 1.8 ks in Ar flow (LLT-mm3).
Fig. 3

XRD patterns of LiCoO2 on different Li0.29La0.57TiO3 solid electrolytes prepared by molten salts: (a) before machining mechanically (bare) and (b) after machining mechanically (LLT-mm3).

Fig. 4

(a) FE-SEM images of fracture surface of LiCoO2 on LLT-mm2, EDX mapping of (b-1) Co-K, (b-2) La-L, (b-3) Ti-K and (c) FE-SEM images of surface of LiCoO2 on LLT-mm2.

3.2 Electrochemical properties

In order to evaluate the electrochemical activities of the LCO generated on the laser-machined LLTO samples, all-solid-state Li secondary batteries were assembled. Figure 5 shows the discharge curves for the LCO prepared from molten salts on various machined LLTO solid electrolytes by changing a current density every two cycles from C/100 to C/10 rate in ASSBs measured at 323 K. For comparison, the charge-discharge curves for the LCO on LLT-bare that is the LLTO before the machining, are attached in Fig. 5(d).

Fig. 5

Charge-discharge curves for LCO prepared from molten salts at 973 K for 3.6 ks (1 h) in on (a) LLT-mm1, (b) LLT-mm2, (c) LLT-mm3 and (d) LLT-bare at various current densities of 0.01, 0.02, 0.04 and 0.1 C measured at 323 K.

At lower current densities the drop of the cell voltage for all investigated LCO samples due to polarization was suppressed, compared with those of LLT-bare. These results may be attributable to an increased contacting area between LCO and LLTO by the laser-machining. Table 3 summarizes the specific capacity and capacity obtained at C/100-rate from Fig. 5, in addition to the estimated surface areas of the laser-machined LLTO samples and the amount of LCO loading. The amount of LCO loading increased 5 to 13 times compared to LLT-bare, in contrast the surface area of LLTO increased 1.8 to 2.6 times. This considerable increase of the amount of LCO loading could be explained by a difference of microstructures of the LCO on the laser-machined LLTO solid electrolytes, depending on machining shape.

Table 3 Summary of results of charge-discharge measurements of LCO on various LLTO solid electrolytes.

The discharge capacity of LLT-mm3 achieved 0.40 mAh cm−2; that of LLT-bare showed 0.032 mAh cm−2. It is well known that a discharge capacity increases with the amount of active electrode material in the conventional liquid Li secondary battery, while in ASSBs that does not lead to increase the capacity directly due to a limited contacting area between electrodes and solid electrolytes.

Nevertheless, the discharge capacity increased linearly with the amount of LCO loading, with keeping the specific capacity from 72 mAh g−1 to 78 mAh g−1. These capacities correspond to from 52% to 56% over 137 mAh g−1 that is the theoretical specific capacity of LCO in investigated voltage range. This means that more than 50% of the LCO generated on various machined LLTO solid electrolytes is active electrochemically regardless of the amount of LCO loading. In addition, the discharge capacity of 60 mAh g−1 was obtained for LLT-mm1 even at a high current density of C/10-rate. These results suggest that the molten salts with fluidity during heating as the starting materials of LCO work effectively to make a favorable interface between LCO and LLTO. However, the discharge capacities for all samples decreased rapidly with an increase in the current density. A further investigation is needed.

4. Conclusion

The surface of LLTO sintered body was processed to various shapes by a picosecond laser with a wavelength of 532 nm. The LLTO crystal phase by the machining remained unchanged. The machining increased the amount of LCO loading and enlarged the discharge capacity in ASSBs. The fraction of the electrochemical active LCO was more than 50% and stayed almost constant regardless of both the shape and surface area of LLTO. These results demonstrated a high electrochemical activity of the interface between LLTO and LCO prepared from molten salts. In addition, these findings provide a promising potential of the process using a liquid-phase sintering by molten salts to construct an electrochemical active interface between oxides.

Acknowledgments

This work was supported by the Japan Science and Technology Agency, Advanced Low Carbon Technology Research and Development Program-Specially Promoted Research for Innovative Next Generation Batteries (JST ALCA-SPRING).

REFERENCES
 
© 2022 Japan Society of Powder and Powder Metallurgy
feedback
Top