Electrochemistry
Online ISSN : 2186-2451
Print ISSN : 1344-3542
ISSN-L : 1344-3542
Articles
Fabrication of LiCoO2 Composite Electrodes for All-solid-state Li Secondary Batteries via Liquid Sintering Using Porous La2/3−xLi3xTiO3 Substrates
Hijiri OIKAWAYuta YOSHIDATakanori YAMAMOTOYoshinori ARACHI
著者情報
ジャーナル オープンアクセス HTML

2022 年 90 巻 6 号 p. 067003

詳細
Abstract

Porous La2/3−xLi3xTiO3 (LLTO) for use as a Li ion conducting framework was prepared on sintered Li0.29La0.57TiO3 bodies using either polyvinylpyrrolidone-assisted sol-gel or screen-printing methods. A mixture of Li and Co salts was subsequently applied to this porous material followed by heating at 700 °C for 1 h to obtain LiCoO2 (LCO)-LLTO composite positive electrodes. These electrodes were subsequently assessed as components of all-solid-state Li secondary batteries (ASSBs). The effect of the amount of LCO loading on the electrochemical performance of these devices was evaluated, based on charge-discharge testing of ASSBs made by inserting Li3.25Ge0.25P0.75S4 between the LLTO and an In-Li alloy negative electrode. The porous LLTO permitted efficient utilization of the active material, and LCO loadings ranging from 1.1 to 6.5 mg cm−2 were possible, with a maximum capacity of 0.65 mAh cm−2. The electrochemical activity of 1 to 2 µm thick LCO layers prepared from molten salts on LLTO was confirmed in this work and LCO was demonstrated to function effectively as a component of a composite positive electrode. This synthesis using liquid sintering with molten salts is a promising approach to forming active electrochemical interfaces between oxides.

1. Introduction

All-solid-state Li secondary batteries (ASSBs) are expected to be adopted as the next generation of storage batteries because they do not present the risk of liquid leakage or combustion. These batteries can also operate at higher potentials and exhibit increased energy densities. Based on achieving the energy and power density values required for practical applications, the use of a lithium anode, a β-Li3PS4 solid electrolyte and a LiNi0.6Co0.2Mn0.2O2 cathode has been proposed.1

Oxide-based ASSB concept could be viable based on predicted manufacturing methods, costs and performance.2,3 To date, oxide-based ASSBs have been fabricated exclusively as thin-film type units, employing vapor deposition processes such as sputtering or pulsed laser deposition (PLD).48 As an example, a LiCoO2 (LCO) membrane with a thickness of 0.2 µm formed using PLD showed good discharge capacity retention through 1000 cycles when combined with Li3PO4 as the solid electrolyte,6 while a 20 µm film thick demonstrated a capacity of 470 µAh cm−2.7 Recently, high-quality (104)-oriented LCO epitaxial thin films have been reported to exhibit exceptional high-rate discharge values exceeding 100,000 C rate and 100 mA cm−2 without any capacity fading.8

Investigations toward bulk-type ASSBs have been actively promoted in order to expand the application and widespread usage of the oxide-based ASSBs.912 Good physical contact has been demonstrated between oxides when using solid electrolytes possessing thermal plasticity, such as Li3BO3-Li2SO4 or Li3BO3-Li2SO4-Li2CO3, as well as with other oxo-acid lithium salt-based compounds.911 In addition, hot-pressing using spark-plasma sintering (SPS) has been shown to provide a superior interface between oxides.12,13 There have been numerous reports concerning garnet-type Li7La3Zr2O12 (LLZO)-based ASSBs. They exhibit exceptional performance at room temperature as a result of the high ionic conductivity of LLZO and the concrete battery designs are also proposed.2,1418 In one study, a combination of Li3BO3 (which has a low melting point) with LCO provided an interfacial resistance of 230 Ω cm2 when in contact with Nb-doped LLZO.14 It has also been reported that porous scaffolds prepared using a freeze-tape-casting method have a garnet-like electrolyte framework. Penetration of a cathode slurry into porous LLZO has been shown to produce a LiNi0.6Mn0.2Co0.2O2 composite electrode with a loading of 4–5 mg cm−2 and a thickness of approximately 120–130 µm.

There have been reports concerning ASSBs using the perovskite-type compound La2/3−xLi3xTiO3 as a solid electrolyte.1922 Following the pioneering fabrication of three-dimensionally ordered macroporous (3DOM) Li0.35La0.55TiO3 using colloidal crystal templating,19 a LiMn2O4 composite positive electrode was found to provide a volumetric capacity of 220 mAh cm−3 for the first time.20 Subsequently, Kotobuki et al. examined the electrochemical performance of LiCoO2 and LiMn2O4 composite electrodes in combination with a porous electrolyte generated in LTTO having a 3DOM structure. The performance of a full cell using Li4Mn5O12 as the negative electrode has also been reported.21,22 However, to the best of our knowledge, the effects of varying the amount of the active electrode materials have not yet been established.

As part of our work concerning bulk-type battery structures, we have focused on the use of ceramic materials to fabricate chargeable/dischargeable interfaces. Specifically, this research has examined the crystal growth of LCO on sintered bodies capable of conducting Li ions. Recently, we reported the electrochemical performance of an LCO film formed on a sintered LLTO body by heating molten salts (serving as the raw material) at 700 °C for 1 h.23 This procedure generated a 1 to 2 µm thick layer of LCO in intimate contact with the LLTO that showed high electrochemical activity. Because LLTO is chemically stable and can be easily handled in air with no special precautions, this material could potentially be used in ASSB half-cells. However, when the thickness of the LCO layer was increased to 14 µm, considerable polarization was observed. Consequently, the present work attempted to prepare various porous LLTO textures on LLTO sintered bodies to increase the contact area with the LCO and to increase the capacity value by generating an LCO layer on the LLTO having a three-dimensional distribution, based on the use of molten salts.

In the research reported herein, various porous LLTO specimens having different surface textures were prepared using sol-gel and screen-printing methods. These specimens were subsequently filled with a mixture of Li and Co molten salts serving as the raw materials for the synthesis of LCO, followed by sintering at 700 °C for 1 h. The resulting LCO-LLTO composites were evaluated as positive electrodes in ASSBs.

2. Experimental

The starting materials for the preparation of La2/3−xLi3xTiO3 using a sol-gel method assisted by polyvinylpyrrolidone (PVP) comprised LiNO3 (>99.9 %, FUJIFILM Wako Pure Chem. Co.), La(NO3)3·6H2O (>99.9 %, FUJIFILM Wako Pure Chem. Co.) and [CH3CH(O-)CO2NH4]2Ti(OH)2 (50 wt% in H2O, Sigma-Aldrich Japan G. K.). The coating solution was made by dissolving these compounds in a 1 : 2 (v/v) H2O : acetic acid mixture, after which a quantity of PVP (K-30, molecular weight = 40,000, FUJIFILM Wako Pure Chem. Co.) was added, followed by stirring for 12 h at 150 rpm. This mixture was subsequently used to coat on a sintered Li0.29La0.57TiO3 body (hereafter denoted as bare-LLTO, 0.5 mm thick ×ϕ10 mm, TOHO TITANIUM Co., Ltd.)24 using a spin-coating device (MS-A100, Mikasa Co. Ltd.), applying a rotation of 5000 rpm for 45 s followed by drying. This coating operation was repeated 10 times. In the case of screen-printing, a 34 : 6 : 60 (w/w/w) Li0.29La0.57TiO3 powder (TP-02F, TOHO TITANIUM Co., Ltd.) : binder (KWE-250 T, Taisei Fine Chemical Co., Ltd.) : polymethylmethacrylate (PMMA; MK-300 in the form of spherical beads, Soken Chemical & Engineering Co., Ltd.) mixture was prepared by stirring the raw materials together for 0.5 h. The preparation of LiCoO2 using a molten salt technique has been previously described in the literature.23 Briefly, the starting materials were LiNO3, LiCl and Co(NO3)2·6H2O. Stoichiometric mixtures of these salts were placed on various porous LLTO samples followed by heating at 700 °C for 1 h in air. X-ray diffraction (XRD) analyses were performed using CuKα radiation to identify crystal phases (XRD-6100, Shimadzu Co., Ltd.). Raman spectra were acquired from the LCO surfaces using laser microscopy (RAMAN Touch, Nanophoton Co.). Field emission – scanning electron microscopy (FE-SEM; JSM-7401F, JEOL Ltd.) observations of the surfaces of machined LLTO samples and analyses by energy dispersive spectroscopy (EDS) to generate elemental maps were conducted at 15.0 kV. The electrochemical properties of the various specimens were examined with an all-solid-state cell tester. Prior to these trials, Au was applied to the LCO side of the device by sputtering to generate a current collector. An In-Li alloy having an electrode potential of 0.62 V vs. Li+/Li was used as the anode electrode. Li3.25Ge0.25P0.75S4 (LGPS) was inserted in the thickness of 0.70 µm between the LLTO and the anode electrode to prevent the reduction of the LLTO by Li. The LGPS was prepared according to a method previously reported by Kanno et al.25 The experimental error associated with the specific capacity values obtained in the present work was in the range of 1–2 mA hg−1. Charge-discharge testing was conducted galvanostatically at 323 K (HJ1001 SD8, HOKUTO DENKO Co.) while applying various current densities. In these trials, a current of 137 mA g−1 was defined as 1 C-rate and tests were performed within the voltage range from 2.0 to 3.6 V, corresponding to 2.6 to 4.2 V vs. Li+/Li. Impedance data were obtained using an impedance analyzer (VSP, Bio-Logic Science Instruments) over the frequency range from 1 MHz to 0.1 Hz and with an applied voltage of 100 mV.

3. Results and Discussion

3.1 Preparation of porous LLTO and LCO composites

Attempts were made to prepare LLTO having a porous texture on sintered Li0.29La0.57TiO3 bodies (bare-LLTO) by a PVP-assisted sol-gel method. This method is able to produce a dense crystalline BaTiO3 film with a thickness of 1–2 µm in close contact with the substrate at relatively low temperatures.26 During this process, PVP molecules undergo electrostatic interactions with metal ions in the solution and an ordered metal-polymer complex is formed that accelerates crystallization at low temperatures. The microstructure of the product is known to be affected by the various preparation parameters, including the concentrations of metal ions and of PVP as well as the drying and heating conditions. This technique was used to form a porous La2/3−xLi3xTiO3 membrane on the bare-LLTO substrate as an alternative to a dense membrane. The gas by pyrolysis of the organic substances contained in the gel during heating is thought to produce pores in the membrane. Preliminary trials were used to determine the optimal preparation conditions for the LLTO phase by this method. In these experiments, LiNO3, La(NO3)3·6H2O and dihydroxybis(ammonium lactato) titanium(IV) were used as the starting materials. An aqueous solution of these compounds was combined with PVP to give a PVP : metal ion molar ratio of 1 : 2, after which the mixture was stirred for 12 h and then sintered at 700, 900 or 1100 °C for 1 or 3 h. Figure 1 presents XRD patterns for LLTO powders prepared by the PVP-assisted sol-gel method and sintered at various temperatures. The specimen heated at 700 °C for 1 h generated broad peaks, whereas these peaks became sharper with increasing sintering temperature, indicating crystal growth of the LLTO. Each peak could be indexed to a tetragonal LLTO phase. The tetragonal lattice parameters for the sample sintered at 1100 °C for 3 h were calculated to be a = b = 3.8718(17) Å and c = 7.7339(65) Å, and these values were consistent with those reported for Li0.33La0.56TiO3 (a = b = 3.872 Å, c = 7.7426 Å).27

Figure 1.

XRD patterns obtained from LLTO powders prepared using a PVP-assisted sol-gel method with heating at (a) 700 °C for 1 h, (b) 900 °C for 1 h, (c) 1100 °C for 1 h, and (d) 1100 °C for 3 h.

Spin-coating was used to fabricate porous LLTO films on bare-LLTO substrates. Figure 2 shows FE-SEM images of the surfaces of various LLTO specimens sintered at 1100 °C for 3 h. These materials contained different proportions of PVP and had been subjected to various preheating temperatures and heating rates. These parameters comprised: sg-a-LLTO, a PVP : metal ion ratio of 1.0, 450 °C, 10 °C min−1; sg-b-LLTO, 0.5, 350 °C, 10 °C min−1 and sg-c-LLTO, 0.5, 350 °C, 4 °C min−1. Depending on these conditions, the microstructure varied significantly whereas the thickness of each film was relatively constant at approximately 2 µm. Both the sg-a-LLTO and sg-b-LLTO contained large grooves with widths of 1–2 µm and lengths of 10–20 µm that were the result of cracking that occurred during heating. In contrast, the sg-c-LLTO had no cracks, although pores and particles with sizes of less than 1 µm were distributed through the entire sample. A well-sintered parts between spherical particles were observed, suggesting a firm framework. These results indicated that LLTO films with pore sizes on the order of 1 µm could be obtained using this PVP-assisted sol-gel method but could contain large cracks exceeding 10 µm depending on the synthesis conditions.

Figure 2.

FE-SEM images of the surfaces of LLTO specimens sintered at 1100 °C for 3 h made using various PVP : metal ion molar ratios, preheating temperatures and heating rates as follows: (a) sg-a-LLTO, 1.0, 450 °C, 10 °C min−1, (b) sg-b-LLTO, 0.5, 350 °C, 10 °C min−1, and (c) sg-c-LLTO, 0.5, 350 °C, 4 °C min−1.

Screen-printing of LTTO powder having an average particle size of 1.2 µm onto the bare-LLTO was employed in order to obtain a thick porous LLTO texture. Two types of slurry were prepared for these trials. These were slurry 1 with an LLTO powder : sintering binder mass ratio of 4 : 6 and slurry 2 with an LLTO powder : PMMA beads (average particle size; 2.8 µm) : sintering binder mass ratio of 3.4 : 0.6 : 6. The suppliers of the binder and PMMA used in this work confirmed that these materials are completely decomposed at 500 and 400 °C, respectively. Figure 3 shows FE-SEM images of the surfaces ((a-s), (b-s), (c-s)) and fracture surfaces ((a-f), (b-f), (c-f)) of various LLTO specimens sintered at 1100 °C using a YSZ plate as a substrate while varying the slurry and sintering time as follows: sp-a-LLTO, slurry 1, 3 h; sp-b-LLTO, slurry 1, 12 h; sp-LLTO, slurry 2, 12 h.

Figure 3.

FE-SEM images of the surfaces ((a-s), (b-s), (c-s)) and fracture surfaces ((a-f), (b-f), (c-f)) of LLTO specimens sintered at 1100 °C using a YSZ plate as a substrate while changing the slurry and sintering time as follows: (a-s), (a-f) sp-a-LLTO, slurry 1, 3 h, (b-s), (b-f) sp-b-LLTO, slurry 1, 12 h, and (c-s), (c-f) sp-LLTO, slurry 2, 12 h.

The particle size for the sp-a-LLTO was determined to be almost the same as that for the raw material before sintering, whereas the sp-b-LLTO particle size was slightly increased to approximately 2 µm. A similar trend was observed in terms of the pore size. Prolonged sintering evidently accelerated the sintering process while retaining the original porosity of the LLTO. However, the apparent porosity estimated based on the mass, dimensions and density of the bare-LLTO for both samples was less than 30 %. As a consequence of adding PMMA beads (as a pore-forming agent) and applying a longer sintering time, a more highly porous LLTO texture (47 % porosity) having a sintered microstructure was obtained. The thickness of the porous material was approximately 20 µm. This porous LLTO (denoted herein as sp-LLTO), in which the pores were 1–2 µm in size, was employed for the following experiments.

After preparing porous LLTO films on bare-LLTO by both the PVP-assisted sol-gel and screen-printing methods, a mixture of Li and Co salts was applied to the porous material followed by heating at 700 °C for 1 h. The maximum mass of LCO that was retained on the substrate was dependent on the apparent porosity of the material. The LCO loadings were determined to be 2.6, 1.9 and 1.1 mg cm−2 for the sg-a-LLTO (porosity; 46 %), sg-b-LLTO (42 %) and sg-c-LLTO (40 %), respectively. These data suggest that LCO may have been generated locally within cracks in the substrates. An attempt was made to fill a thicker porous body to increase the LCO loading but the LCO was found only to be deposited on the surface of this substrate and so the loading was not improved. The LCO loading on the sp-LLTO ranged from 1.3 to 6.5 mg cm−2 because this material had a porous surface structure with a thickness of 20 µm (that is, 10 times thicker than that obtained using the PVP-assisted sol-gel method) and also contained numerous voids. XRD patterns obtained from the LCO samples are presented in Fig. 4. All peaks in these patterns could be assigned to a standard high-temperature LiCoO2 phase except for those derived from the LLTO substrate and the sample-fixing agent.

Figure 4.

XRD patterns obtained from (a) an LLTO substrate (bare-LLTO), (b) LiCoO2 on sg-c-LLTO and (c) LiCoO2 on sp-LLTO prepared from molten salts at 700 °C for 1 h.

3.2 Electrochemical properties

ASSBs incorporating positive electrodes comprising LCO contained in sg-c-LLTO or sp-LLTO were assembled, and constant current charge/discharge measurements were carried out at 50 °C. Charging was at a C/25 rate up to 3.6 V followed by the application of a constant voltage for 12 h. Discharging of the sg-c-LLTO was at C/100, C/50, C/25 or C/10 rates while the sp-LLTO was discharged in different order, C/50, C/25, C/10 and C/100 rates down to 2.0 V. Two cycles were monitored for each C-rate. The discharge curves acquired from the sg-c-LLTO with an LCO loading of 1.1 mg cm−2 and from the sp-LLTO with a loading of 1.3 mg cm−2 are shown in Figs. 5a and 5b, respectively. The discharge capacity of the sg-c-LLTO at the C/100 rate was 93 mAh g−1 while the value for the sp-LLTO was 105 mAh g−1. Thus the LCO utilization values were 71 % and 77 %, respectively, relative to the theoretical capacity of 137 mAh g−1. Both cells maintained a high discharge capacity even when the rate was increased to C/25, although the sg-c-LLTO cell exhibited a remarkable I × R drop at the C/10 rate. Reduction of the average discharge voltage was suppressed, indicating an increase in the contact area between the LCO and the LLTO serving as the ion conductor, and the reduced overvoltage resulted in more efficient utilization of the LCO. These results confirmed that LCO contained in porous LLTO was able to provide a composite positive electrode for use in an ASSB.

Figure 5.

Discharge curves for ASSBs based on (a) a LiCoO2 electrode (1.1 mg cm−2) on sg-c-LLTO and (b) a LiCoO2 electrode (1.3 mg cm−2) on sp-LLTO prepared from molten salts by changing the current density every two cycles with heating at 50 °C.

The effect of the amount of LCO loading on the electrochemical performance was also evaluated. These trials used sg-a-LLTO and sg-b-LLTO specimens with loadings of 2.6 and 1.9 mg cm−2, respectively, along with sp-LLTO samples having loadings of 4.8 and 6.5 mg cm−2. After performing the electrochemical tests described above, the discharge curves for LCO-LLTO composites synthesized by the sol-gel and screen-printing methods acquired at C/100 and C/10 rates were extracted, and are shown in Figs. 6a and 6b, respectively. Figure 6a shows that the discharge capacities at both the C/100 and C/10 rates decreased with increasing amount of LCO loading, indicating decreased utilization of the LCO. The potential plateau around 3.3 V is also seen to have disappeared at the C/10 rate. The ohmic resistances estimated from the I × R drops were in the range of 16–17 kΩ cm−2. In addition to the low ionic conductivity of the porous body, inactive LCO aggregates were evidently localized in cracks larger than 10 µm in the porous structure. As a result, the LCO was used less effectively. The ionic conductivity of an LLTO film produced using the sol-gel method has been reported to be 4.5 × 10−6 S cm−1, which is two to three orders of magnitude lower than that of the bulk material.28 Therefore, it is likely that the porous texture generated by the sol-gel method has an inherently high electrical resistance related to ion transport. The data in Fig. 6b indicate that the three discharge curves obtained from the sp-LLTO at the C/100 rate almost overlapped and each showed the same discharge capacity of 105 mAh g−1, corresponding to a utilization of 77 % and a maximum capacity of 0.68 mAh cm−2. The potential plateau was maintained at the C/10 rate regardless of the LCO loading, although an I × R drop was observed and the discharge capacity decreased. The estimated ohmic resistance was in the range of 2.8–3.6 kΩ cm−2 and so was less than 20 % of the values observed in the case of materials made using the PVP-assisted sol-gel method. Although the sp-LLTO composite electrode was approximately 10 times thicker than the sg-LLTO electrode, the ohmic loss for the former was very low. The strong bonding between the LLTO particles and the porous sp-LLTO likely enhanced Li ion conduction in this material. It is also interesting to note that the sp-LLTO composite electrode, which showed a suppressed I × R drop, maintained a high proportion of the original LCO electrochemical activity over a wide range of LCO loadings. These results demonstrate that a 1 to 2 µm thick LCO thick layer obtained from molten salts and closely attached to an LLTO substrate exhibited electrochemical activity even in a porous body having 1 to 2 µm pores.

Figure 6.

Discharge curves for ASSBs with LiCoO2 specimens having varying LCO loadings: (a) 1.1 (sg-c-LLTO), 1.9 (sg-b-LLTO) and 2.6 mg cm−2 (sg-a-LLTO), and (b) 1.3, 4.8 and 6.5 mg cm−2 in sp-LLTO prepared from molten salts at a current density of C100 and C10 rates with heating at 50 °C.

The distribution of LCO in the composite electrode made using the sp-LLTO was examined by EDS mapping of the fracture surface, as shown in Fig. 7. Co was found to be located in various regions of the specimen, associated with both La and Ti. These data confirm that LCO was generated in the pores of the LLTO. The Co appears to have been distributed to a depth of approximately 40 µm below the electrode surface, but in reality the depth was only approximately half this value, with the apparent error due to the fracture in this sample. The Raman spectrum of the surface of the LCO composite electrode made using sp-LLTO is presented in Fig. 8. In addition to the peaks expected for a standard high-temperature LCO phase, two peaks at 450 and 678 cm−1 attributed to a low-temperature LCO phase and a Co3O4 phase appeared.29 Although these phases were not detectable in the corresponding XRD patterns, they were evidently present in these samples and suggest that changes in the stoichiometry of the LiCoO2 affected its utilization. It appears that liquid sintering helped to construct an electrochemically active interface between the oxides by virtue of the fluidity of the molten salt, although it may be difficult to precisely control the chemical composition of the product.

Figure 7.

(a) FE-SEM image of the fracture surface of the LiCoO2/sp-LLTO interface and EDS element mappings showing the distributions of (b-1) Co, (b-2) La and (b-3) Ti.

Figure 8.

Raman spectra acquired from sp-LLTO loaded with LiCoO2 prepared from molten salts.

Impedance assessments were conducted using an ASSB incorporating LCO on the sp-LLTO at a loading of 2.9 mg cm−2 in conjunction with various electrode potentials. The resulting Nyquist plots are shown in Fig. 9a while those obtained from an ASSB made using the bare-LLTO with a loading of 0.89 mg cm−2 before charge-discharge testing are provided in Fig. 9b for comparison. The depressed semicircle in the high-frequency region around 10 kHz was found not to undergo any significant change regardless of the potential, suggesting that ohmic resistance was dominant (in agreement with the results obtained from charge-discharge measurements). Compared with the bare-LLTO, the semicircle generated by the porous LLTO was noticeably smaller. It is also evident that the values in the low-frequency region (below 200 Hz) greatly decreased with increasing potential. This effect is believed to be related to the charge transfer resistance between the LCO and LLTO. The interfacial resistance between LLTO and LCO connected by SPS due to interdiffusion has been reported to be 105 Ω.30 However, it is possible that the present interfacial resistance resulted from crystal growth of the LCO, which lowered the resistance of the material. A detailed impedance analysis is currently underway in our laboratory.

Figure 9.

Nyquist plots obtained from (a) In-Li/LGPS/LLTO/LCO-sp-LLTO at various electrode potentials and (b) In-Li/LGPS/LLTO/LCO-bare-LLTO before charge-discharge testing.

The effects of the retention of discharge capacity observed during the first cycle at each C-rate on various C-rates for each of the LCO composites are summarized in Figs. 10a and 10b, in addition to the data obtained from the bare-LLTO for comparison. At LCO loadings of up to 1.9 mg cm−2, the retention exceeded 80 % at all C-rates and the rate performance was slightly improved by use of a porous texture. In contrast, at loadings greater than 1.9 mg cm−2, the retention decreased linearly with increasing C-rate. It is interesting to note that the results obtained with loadings of 4.8 and 6.5 mg cm−2 on the sp-LLTO showed almost the same rate dependence. These data can likely be attributed to the low ohmic resistance of the ASSBs when using this analytical method, as was also observed in the discharge curves and Nyquist plots. In the case of a 6.7 µm thick LCO film generated by PLD on Li3PO4, there is an upper limit to the rate of chemical diffusion of Li, increasing current density.7 Thus, further reductions in the cell resistance are required. Figure 10c shows the relationship between the retention of the discharge capacity and the cycle number for the sp-LLTO with a loading of 1.3 mg cm−2. A previous study of a thin LCO film deposited on LLTO by sputtering demonstrated that amorphous LLTO contributed to the stress relaxation related to the expansion and contraction associated with the charge and discharge of the LCO.31 The present composite having a porous texture was relatively stable during cycling, possibly because it contained numerous void spaces.

Figure 10.

C-rate dependency of the retention of discharge capacity for LiCoO2 samples with varying LCO loadings: (a) 1.1 mg cm−2 on sg-c-LLTO, 1.9 mg cm−2 on sg-b-LLTO and 2.6 mg cm−2 on sg-a-LLTO and (b) 1.3, 4.8 and 6.5 mg cm−2 on sp-LLTO, prepared from molten salts in ASSBs at 50 °C. (c) Relationship between the retention of discharge capacity and the cycle number for sp-LLTO with an LCO loading of 1.3 mg cm−2 in an ASSB.

4. Conclusion

Sintered Li0.29La0.57TiO3 bodies having porous LTTO textures were formed by either PVP-assisted sol-gel or screen-printing techniques, and then coated with LCO prepared using molten salts. Electrochemical measurements showed that the porous LLTO fabricated by screen-printing exhibited efficient utilization of the active material in conjunction with LCO loadings ranging from 1.1 to 6.5 mg cm−2 and achieved a maximum capacity of 0.65 mAh cm−2. The rate performance was also slightly improved when using this composite electrode, and a 1 to 2 µm thick layer of LCO closely attached to the LLTO was determined to provide superior electrochemical activity. It is evident that liquid sintering using molten salts has the potential to produce electrochemically active interfaces 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). The authors acknowledge the providing the LLTO samples by TOHO TITANIUM, Japan.

CRediT Authorship Contribution Statement

Hijiri Oikawa: Data curation (Lead), Investigation (Lead), Writing – original draft (Lead)

Yuta Yoshida: Data curation (Equal), Investigation (Equal)

Takanori Yamamoto: Investigation (Equal)

Yoshinori Arachi: Conceptualization (Lead), Funding acquisition (Lead), Supervision (Lead), Writing – original draft (Equal), Writing – review & editing (Lead)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Footnotes

Y. Yoshida: ECSJ Student Member

Y. Arachi: ECSJ Active Member

References
 
© The Author(s) 2022. Published by ECSJ.

This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium provided the original work is properly cited. [DOI: 10.5796/electrochemistry.22-00043].
http://creativecommons.org/licenses/by/4.0/
feedback
Top