JP7684286B2 - Lithium-ion conductive glass ceramics - Google Patents

Description

The present invention relates to lithium ion conductive glass ceramics.

2. Description of the Related Art In recent years, lithium-ion secondary batteries, which have high energy density and are capable of being charged and discharged, have been widely used for applications such as power sources for electric vehicles and mobile phone terminals.
Many of the lithium ion secondary batteries currently available on the market use a liquid electrolyte (electrolytic solution) to achieve high energy density, and the electrolyte is made by dissolving a lithium salt in an aprotic organic solvent such as a carbonate ester or a cyclic ester.

However, in lithium-ion secondary batteries that use liquid electrolytes (electrolytic solutions), there is a risk of the electrolyte leaking. In addition, organic solvents that are commonly used in electrolytes are volatile and flammable, which poses safety concerns.

Therefore, it has been proposed to use a solid electrolyte as the electrolyte for lithium-ion secondary batteries instead of a liquid electrolyte (electrolytic solution) such as an organic solvent. Furthermore, development of an all-solid-state secondary battery that uses a solid electrolyte as the electrolyte and has other components all made of solids is underway.
Representative properties required for the solid electrolyte of an all-solid-state secondary battery include lithium ion conductivity, reduction resistance, stability during firing, and stability over time.

For example, Patent Document 1 discloses a solid electrolyte containing a glass ceramic containing a crystal phase of Li1 _{+x+zMx (} Ge1 _{- yTiy} ) _{2- xSizP3 - zO12} (where 0≦x≦0.8, 0≦y≦1.0, 0≦z≦0.6, M=one or more selected from Al and Ga) as a solid electrolyte to be interposed between a positive electrode containing an active material and a negative electrode containing an active material. In addition, Non-Patent Document ₁ discloses NASICON type _{Li1.15Y0.15Zr1.85} ( _PO4 ) ₃ having high lithium _ion conductivity at room temperature for use as a solid electrolyte.

JP 2009-181921 A

Journal of Power Sources 240 (2013) 50-53

However, since the glass ceramic described in Patent Document 1 contains Ti as a constituent component, its reduction resistance at 25°C is about 2.5 V (vs. Li), and when used as a solid electrolyte, there is a problem that low _{- potential} negative electrode active materials such as _Li4Ti5O12 and _TiO2 cannot be used.

According to the inventors' investigation, the solid electrolyte described in Non-Patent Document 1 has been improved in reduction resistance at 25° C. to 1.5 V (vs. Li) or less, and has been reported to have a relatively high lithium ion conductivity of 0.7×10 ⁻⁴ S·cm ⁻¹ at 25° C. However, in order to suppress grain growth, it is necessary to use a current sintering method (SPS) in which electricity is applied at high temperatures and pressure is applied while sintering, which is an industrially significant problem. In addition, this results in a solid electrolyte that contains almost no amorphous matter (mostly crystalline phase), making it difficult to obtain glass ceramics. Note that when this solid electrolyte is produced without using the current sintering method, grain growth is significant, and the grain boundary resistance (resistance to ion conduction occurring at the contact interface between particles) of the obtained solid electrolyte is high, and the lithium ion conductivity is less than 1.0×10 ⁻⁵ S·cm ⁻¹ .

Therefore, the present invention aims to provide a lithium ion conductive glass ceramic that can be easily produced industrially, has low grain boundary resistance, high lithium ion conductivity and high reduction resistance, and can be used as a solid electrolyte for all-solid-state secondary batteries, etc.

In order to solve the above problems, the present inventors have conducted extensive research and have found that _{a lithium} ion conductive glass ceramic containing, in mole percent on an oxide basis, 10-20% _Li2O , 30-40% _P2O5 , 40-50% _{ZrO2, 0-4% Y2O3 , 0-3} % _Al2O3 , and 0-2% _GeO2 , which contains a crystalline phase having a rhombohedral NASICON structure and in which the a-axis lattice constant of this crystalline phase, as identified by X-ray diffraction and Rietveld analysis, is 8.872 Å or more, can be easily produced industrially, and has low grain boundary resistance, high lithium ion conductivity, and high reduction resistance, thereby completing the present invention.

That is, the present invention relates to the following (1) to (10).
(1) A _lithium ion conductive glass-ceramic containing, in mole percent on an oxide basis, 10-20% Li2O, 30-40% _P2O5 , 40-50% _ZrO2 , 0-4% _Y2O3 , 0-3% _Al2O3 , and 0-2% _GeO2 , and containing _a rhombohedral _NASICON structure crystal phase, the a-axis lattice constant of which, as identified by X-ray diffraction and Rietveld analysis, is 8.872 _Å or more.
(2) The lithium ion conductive glass-ceramics according to (1), in which the sum of the masses of crystalline components identified by X-ray diffraction and Rietveld analysis is less than 97 mass% of the total mass.
(3) The lithium ion conductive glass-ceramics according to (1) or (2), wherein the crystal phase includes a LiZr ₂ (PO ₄ ) ₃ α phase and/or a LiZr ₂ (PO ₄ ) ₃ α' phase.
(4) The lithium ion conductive glass-ceramics according to (3), wherein the mass of the LiZr ₂ (PO ₄ ) ₃ α phase identified by X-ray diffraction and Rietveld analysis is 80 mass % or more based on the total mass.
(5) The lithium ion conductive glass ceramic according to any one of (1) to (4), containing 0.1 to 5% of a SiO ₂ component, in mole percent based on oxide.
(6) The lithium ion conductive glass ceramic according to (5), wherein the valence of Si contained in the entire lithium ion conductive glass ceramic is 3.5 or more and less than 3.9.
(7) The lithium ion conductive glass ceramic according to (5) or (6), wherein the coordination number of O to Si contained in the lithium ion conductive glass ceramic is 5 or more and 7 or less in the entire lithium ion conductive glass ceramic.
(8) The lithium ion conductive glass ceramic according to any one of (1) to (7), which is a powder having a lithium ion conductivity at 25° C. of 1.0×10 ⁻⁵ S·cm ⁻¹ or more.
(9) The lithium ion conductive glass ceramic according to any one of (1) to (7), which is a substrate having a lithium ion conductivity at 25° C. of 7.0×10 ⁻⁵ S·cm ⁻¹ or more and a thickness of 300 μm or less.
(10) The lithium ion conductive glass ceramic according to (9), wherein the maximum particle size of the particles in the outermost particle layer of the substrate is 30 μm or less, and the average particle size of the particles is 15 μm or less.

According to the present invention, it is possible to obtain a lithium ion conductive glass ceramic that can be easily manufactured industrially, has low grain boundary resistance, high lithium ion conductivity, and high reduction resistance. This lithium ion conductive glass ceramic is a solid electrolyte with a wide potential window on the low potential side due to its high reduction resistance, and can therefore be used in combination with low potential electrode active materials and high potential electrode active materials, resulting in a high voltage battery. Therefore, it can be suitably used as a solid electrolyte for all-solid-state secondary batteries, seawater batteries, and the like.

1 is a graph showing the impedance measurement results (Cole-Cole Plot) of the solid electrolyte pellets of Example 2. 1 is a secondary electron image of an outermost particle layer in a solid electrolyte pellet of Comparative Example 1 (a photograph substituted for a drawing). 1 is a secondary electron image of an outermost particle layer in a solid electrolyte pellet of Example 1 (a photograph substituted for a drawing). 1 is a secondary electron image of an outermost particle layer in a solid electrolyte pellet of Example 2 (a photograph substituted for a drawing). 1 is a graph showing the relationship between the a-axis lattice constant of the LiZr ₂ (PO ₄ ) ₃ α-phase and the lithium ion conductivity (S·cm ⁻¹ ) in the solid electrolyte pellets of Example 1 (initial product, after one year) and Example 2 (initial product, after one year). 1 shows cyclic voltammograms of the solid electrolyte pellet of Example 1 (dotted line) and the solid electrolyte pellet of Example 2 (solid line). 1 shows the results of XAFS analysis of the solid electrolyte powder and SiO ₂ powder of Example 2. 1 shows Si K-edge EXAFS data (normalized by absorption edge intensity) for the solid electrolyte powder and SiO ₂ powder of Example 2. 1 shows Si K-edge EXAFS vibration function data ( ^k2 weighted) for the solid electrolyte powder and SiO2 _powder of Example 2. 1 is a graph showing the effect of the ratio of Y on the lithium ion conductivity (S·cm ⁻¹ , solid line) and density (g/cm ³ , dotted line) in the solid electrolyte pellets of Examples 2 to 6. 1 is a graph showing the effect of the Si ratio on the lithium ion conductivity (S·cm ⁻¹ , solid line) and density (g/cm ³ , dotted line) in the solid electrolyte pellets of Examples 1 to 2 and 7 to 9. 1 is a graph showing the effect of the a-axis lattice constant (Å) of the _LiZr2 (PO4) ₃ α-phase, which is changed by varying the Y ratio or the Si ratio, on the lithium ion conductivity (S·cm−1) in the solid electrolyte pellets of Examples 3 to 6 (open plots) and Examples 1 to 2 _{and 7} to ⁹ (filled plots). 1 is a graph showing the impedance measurement results (Cole-Cole Plot) of the solid electrolyte substrate of Example 11. 13 is a secondary electron image of an outermost particle layer in a solid electrolyte substrate of Example 11 (a photograph substituted for a drawing). 13 is a diagram showing a state in which particle size measurement is being performed from a secondary electron image of the outermost particle layer in the solid electrolyte substrate of Example 11 (a photograph as a substitute for a drawing).

The present invention will now be described.
The present invention relates to a _lithium ion conductive glass-ceramic containing, in mole percent on an oxide basis, 10 to 20% _Li2O , 30 to 40% _P2O5 , 40 to 50% _ZrO2 , 0 to 4% _Y2O3 , 0 to ₃ % _Al2O3 , _and 0 to 2% _GeO2 , and containing a crystalline phase with a rhombohedral NASICON structure, the a-axis lattice constant of this crystalline phase being 8.872 Å or more as identified by X-ray diffraction and Rietveld analysis.
Hereinafter, this may be referred to as the "glass ceramic of the present invention."

Here, in the present invention, the term "glass ceramics" refers to a material obtained by heat-treating a raw glass to cause precipitation of a crystalline phase, and includes the crystalline phase formed by the heat treatment and an amorphous phase (non-crystalline phase). In other words, it is a mixture of ceramics and glass.
The content of each component contained in the glass ceramic of the present invention is expressed in terms of mol% based on oxide unless otherwise specified. The composition expressed in "mol% based on oxide" refers to a composition in which each component contained in the glass ceramic of the present invention is expressed with the total number of moles of the generated oxides being 100 mol%, assuming that the oxides, composite salts, metal fluorides, etc. used as raw materials for the glass ceramic of the present invention are all decomposed and converted to oxides during melting.

<Components>
First, each component constituting the glass ceramic of the present invention will be described in detail.

The _Li2O component is an essential component for imparting lithium ion conductivity to the glass ceramic of the present invention. Therefore, the lower limit of the content of the _Li2O component is 10%, preferably 12%, and more preferably 14%. On the other hand, since the chemical durability of the glass ceramic of the present invention can be increased to improve the morphological stability when used as a solid electrolyte layer, the upper limit of the content of the _Li2O component is 20%, preferably 18%, and more preferably 16%.

The _P2O5 component is _an essential component necessary for forming a crystal phase having a rhombohedral NASICON structure in the glass ceramics of the present invention. Therefore, the content of the _P2O5 component is set to 30%, preferably 33%, more preferably 35% as the lower limit. On the other hand, the content of the _P2O5 component is set to 40%, preferably 39%, more preferably 38%, as the content of the _P2O5 component can suppress the formation of other crystal phases and make it difficult to reduce the lithium _ion conductivity of the formed crystal phase.

The _ZrO2 component is an essential component that can make the crystal phase having lithium ion conductivity in the glass ceramic of the present invention less susceptible to decomposition due to reduction. Therefore, the content of the _ZrO2 component is set to 40%, preferably 42%, more preferably 44% as the lower limit. On the other hand, since it can easily form a crystal phase having a rhombohedral NASICON structure, the content of the _ZrO2 component is set to 50%, preferably 48%, more preferably 47% as the upper limit.

In addition, the molar ratio of the P ₂ O ₅ component to the Li ₂ O component (P ₂ O ₅ component/Li ₂ O component) in the glass ceramic of the present invention is more preferably 2.0 to 3.0, and the molar ratio of the ZrO ₂ component to the Li ₂ O component (ZrO ₂ component/Li ₂ O component) is more preferably 2.5 to 3.5.

The _Y2O3 component is _an optional component that can adjust the lithium ion conductivity of the crystalline phase in the glass ceramic of the present invention and can adjust the mechanical strength and size of the crystalline phase. Therefore, the content _of the _Y2O3 component is preferably 0.1%, more preferably 0.5%, even more preferably 1%, and even more preferably 1.2% as the lower limit. On the other hand, since it is possible to suppress the formation of other crystalline phases and make it difficult to reduce the lithium ion conductivity of the formed crystalline phase, the content of the _Y2O3 component is preferably ₄ % as the upper limit, more preferably 3.5%, even more preferably 3.2%, even more preferably 3%, even more preferably 2.8%, and even more preferably 2.5% as the upper limit.
The molar ratio of the Y ₂ O ₃ component to the Li ₂ O component (Y ₂ O ₃ component/Li ₂ O component) in the glass ceramics of the present invention is more preferably 0.05 to 0.26, and even more preferably 0.05 to 0.20.

The _Al2O3 component, like _{the Y2O3} component, is _an optional component that can adjust the lithium ion conductivity of the crystal phase in the glass ceramic of the present invention and can adjust the mechanical strength and size of the crystal phase. Therefore, the content of the _Al2O3 component is preferably 0.1%, more preferably 0.5%, even more preferably 1%, and even more preferably 1.2% as the lower limit. Also, like _{the Y2O3} component _, this can suppress the formation of other crystal phases and make it difficult to reduce the lithium ion conductivity of the formed crystal phase, so the content of the _Al2O3 component is preferably ₃ % as the upper limit, more preferably 2.8%, and even more preferably 2.5% as the upper limit.

The _GeO2 component is an optional component that promotes the crystallization of the glass ceramics of the present invention. Therefore, the content of the _GeO2 component is preferably 0.1%, more preferably 0.5%, and even more preferably 1% as the lower limit. On the other hand, the content _of the GeO2 component is preferably ₂ % or less, and more preferably _1.5 % or less, because the lithium ion conductivity of the crystal phase formed in the same manner as the _Y2O3 component and the Al2O3 component is less likely to decrease, and because the _GeO2 component is easily coexistent with the Zr component.

The _SiO2 component is an optional component that can increase the mechanical strength of the glass ceramic of the present invention and can improve the lithium ion conductivity of the glass ceramic of the present invention by partially replacing the _P2O5 component. Therefore, the content of _{the SiO2} component is preferably set to 0.1%, more preferably 0.5%, and even more preferably 0.7% as the lower limit. On the other hand, since it is possible to easily form a desired crystal phase and the crystals are easily adjacent to each other, and the decrease in lithium ion conductivity can be suppressed, the content of the _SiO2 component is preferably set to 5%, more preferably 4%, more preferably 3%, and even more preferably 2.5% as the upper limit.
The molar ratio of the _SiO2 component to the _Li2O component ( _SiO2 component/ _Li2O component) in the glass ceramics of the present invention is more preferably 0.007 to 0.17, and even more preferably 0.03 to 0.17.

The CaO component and the MgO component are optional components that can increase the lithium ion conductivity by incorporating a large amount of Li into the crystal phase through a balance of valences. Therefore, it is preferable that the lower limit of the content of the CaO component and the content of the MgO component is preferably 0.5%, more preferably 1%, and even more preferably 2%. On the other hand, since it is possible to suppress the decrease in the lithium ion conductivity of the glass ceramic of the present invention, it is preferable that the upper limit of the content of the CaO component and the content of the MgO component is preferably 5%, more preferably 4%, and even more preferably 3%.

The Sc ₂ O ₃ component and the Ga ₂ O ₃ component are optional components that can adjust the lithium ion conductivity of the crystal phase in the glass ceramic of the present invention, as well as the Y ₂ O ₃ component and the Al _{2 O} 3 component, and can adjust the size of the crystal phase. Therefore, the content of the Sc ₂ O ₃ component and the content of the Ga ₂ O ₃ component are preferably set to a lower limit of 0.1%, more preferably 0.5%, and even more preferably 1%. In addition, like the Y ₂ O ₃ component and the Al ₂ O ₃ component, this also suppresses the formation of other crystal phases and makes it difficult to reduce the lithium ion conductivity of the formed crystal phase, so that the content of the Sc ₂ O ₃ component and the content of the Ga ₂ O ₃ component are preferably set to an upper limit of 2%, more preferably 1.5%.

The _SnO2 component is an optional component that promotes the crystallization of the glass ceramic of the present invention, similar to the _GeO2 component. Therefore, the content of the _SnO2 component is preferably 0.1%, more preferably 0.5%, and even more preferably 1% as the lower limit. On the other hand, since the lithium ion conductivity of the formed crystal phase can be made difficult to decrease, the content of the _SnO2 component is preferably 2%, more preferably 1.5% as the upper limit.

Furthermore, the glass ceramics of the present invention may contain inorganic components including boron (B) or fluorine (F).

On the other hand, in the glass ceramics of the present invention, it is preferable to reduce the titanium (Ti) content as much as possible (for example, less than 1%, or even less than 0.1%), and it is more preferable to not contain Ti. This is because reducing the Ti component can suppress the reduction resistance from decreasing. In addition, it is also preferable to reduce not only Ti but also transition metal components such as niobium (Nb), vanadium (V), and nickel (Ni) as much as possible, and it is more preferable to not contain them at all.

Furthermore, in the glass ceramics of the present invention, it is preferable to reduce the content of sulfur (S) components as much as possible, and it is more preferable to not contain any. This is because reducing the S component can reduce the possibility of generating harmful gases such as hydrogen sulfide in all-solid-state secondary batteries. In addition, in order to avoid a decrease in lithium ion conductivity, it is preferable to reduce the content of alkali metals other than Li (Na, K, etc.) as much as possible, and it is more preferable to not contain any.

<Crystal phase, etc.>
Next, the composition of the crystal phase contained in the glass ceramic of the present invention and the composition of the phases other than this crystal phase will be described in detail.

The glass ceramic of the present invention contains a predetermined amount of each of the above-mentioned components, and further contains a crystal phase having a rhombohedral NASICON structure composed of at least a portion of each of these components. It is preferable that all of the crystal phases contained in the glass ceramic of the present invention have a rhombohedral NASICON structure, but other lithium ion conductive crystal phases (e.g., LISICON type, perovskite type, garnet type, etc.) may be partially contained. Even in this case, of all the crystal phases contained in the glass ceramic of the present invention, the crystal phase having a rhombohedral NASICON structure is preferably 80 mass% or more, more preferably 90 mass% or more, and even more preferably 95 mass% or more.

The glass ceramics of the present invention, which includes the above-mentioned crystal phase having a rhombohedral NASICON structure, have an a-axis lattice constant of 8.872 Å or more, as identified by X-ray diffraction (XRD) and Rietveld analysis. The a-axis lattice constant of the crystal phase is elongated to a predetermined value or more, resulting in a glass ceramic with high lithium ion conductivity. The a-axis lattice constant is preferably 8.874 Å or more, more preferably 8.876 Å or more, and even more preferably 8.878 Å or more. The upper limit is preferably 8.892 Å or less, more preferably 8.889 Å or less, and even more preferably 8.888 Å or less.

Furthermore, the glass ceramics of the present invention containing the above-mentioned rhombohedral NASICON-type crystal phase preferably have a mass sum of the crystal components (crystal phase components) identified by X-ray diffraction (XRD) and Rietveld analysis of less than 97% by weight relative to the total mass of the glass ceramics of the present invention. In other words, the mass of the components of the unspecified phase (phase other than the crystal phase) calculated by the difference between the total mass of the glass ceramics of the present invention and the mass sum of the crystal components identified by X-ray diffraction (XRD) and Rietveld analysis is preferably 3% by mass or more relative to the total mass of the glass ceramics of the present invention. This is because it is possible to suppress the increase in grain boundary resistance due to overgrowth of particles, and the extension of the a-axis lattice constant of the crystal phase can be stabilized. The unspecified phase mainly contains an amorphous phase. It is more preferable that the mass sum of the crystal components is less than 95% by weight relative to the total mass of the glass ceramics of the present invention, that is, it is more preferable that the mass of the components of the unspecified phase is 5% by mass or more. The lower limit of the sum of the masses of the crystal components is not limited, but is preferably 80% by weight or more, and more preferably 85% by weight or more, based on the total mass of the glass ceramic of the present invention.

The above-mentioned X-ray diffraction (XRD) was performed using a powder X-ray diffractometer (D8 DISCOVER, manufactured by Bruker) by mixing the same mol% amount of ZnO as the initial sample amount as a standard sample and using CuKα radiation.

In addition, the above-mentioned Rietveld analysis is performed using the Rietveld analysis software "Z-Rietveld code" using the XRD data measured under the above-mentioned conditions.

The crystal phase having a rhombohedral NASICON structure contained in the glass ceramics of the present invention preferably contains a crystal phase of Li1 _{+x+yYxZr2 -xSiyP3 - yO12 (} 0≦ _x <2, 0≦y<3) because this further improves lithium ion conductivity at room temperature, and more preferably contains a _LiZr2 ( _PO4 ) ₃ α phase (high temperature phase) and/or a _LiZr2 ( _PO4 ) ₃ α' phase (low temperature phase). The crystal phase of the rhombohedral NASICON structure may be composed of _LiZr2 ( _PO4 ) _3α phase (high temperature phase) and/or _LiZr2 ( _PO4 ) _3α ' phase (low temperature phase), and may contain other phases of _LiZr2 ( _PO4 ) ₃ (β phase, γ phase, etc.) in addition to the α phase and α' phase. If the _LiZr2 ( _PO4 ) _3α phase, whose a-axis lattice constant is within the above-mentioned range, is contained in an amount of 80 mass % or more, and further 85 mass % or more, based on the total mass of the lithium ion conductive glass-ceramics of the present invention, the stability over time will be more excellent, which is very preferable. Here, the α phase and α' phase are CSD numbers 97658 (α-LiZr ₂ (PO ₄ ) ₃ ) and 89456 (α'-LiZr ₂ (PO ₄ ) ₃ ) of the Cambridge Crystal Structure Database CSD-System.

Furthermore, when the glass ceramic of the present invention contains _SiO2 component, the valence of Si contained in the entire glass ceramic of the present invention (crystal phase and unspecified phase) is more preferably 3.5 or more and less than 3.9, and even more preferably 3.6 or more and 3.8 or less, because the stability of lithium ion conductivity and the mechanical strength of the crystal structure are more excellent.

In addition, when the glass ceramic of the present invention contains a _SiO2 component, the coordination number of O to the contained Si in the entire glass ceramic of the present invention (crystal phase and unspecified phase) is more preferably 5 to 7, and further preferably 6. This is also because the stability of lithium ion conductivity and the mechanical strength of the crystal structure are more excellent.

The above-mentioned Si valence is measured by X-ray absorption fine structure (XAFS).
The coordination number of O to Si is measured by extended X-ray absorption fine structure (EXAFS).

<Form>
Next, the form of the glass ceramic of the present invention will be described in detail.

The form of the glass ceramic of the present invention is not particularly limited, but when used as a solid electrolyte layer or electrode layer of an all-solid-state secondary battery, or an electrode layer or partition layer (separator) of a seawater battery, it is preferable that the glass ceramic of the present invention is in the form of a powder or a substrate. In particular, when the glass ceramic of the present invention is mixed with an electrode active material to form an electrode layer, the glass ceramic of the present invention is preferably in the form of a powder.

The glass ceramic of the present invention has high lithium ion conductivity and high reduction resistance, but when the glass ceramic of the present invention is in the form of a powder, for example, the lithium ion conductivity at 25° C. is preferably 1.0×10 ⁻⁵ S·cm ⁻¹ or more, more preferably 3.0×10 ⁻⁵ S·cm ⁻¹ or more, even more preferably 6.0×10 ⁻⁵ S·cm ⁻¹ or more, and even more preferably 7.0×10 ⁻⁵ S·cm ⁻¹ or more, and the reduction resistance at 25° C. is preferably 1.5 V (vs. Li: potential relative to lithium) or less, more preferably 1.0 V (vs. Li) or less. When the lithium ion conductivity at 25° C. is less than 1.0×10 ⁻¹⁰ S·cm ⁻¹ , lithium ion conduction does not substantially occur. Furthermore, although not limited thereto, the lithium ion conductivity of this powder may be 1.0×10 ⁻² S·cm ⁻¹ or less at 25° C., and the reduction resistance may be 0.3 V (vs. Li) or more at 25° C. Furthermore, it is preferable that the particles contained in this powder have a maximum particle size of 30 μm or less and an average particle size of 15 μm or less.

In addition, when the glass ceramic of the present invention is a substrate, it is preferable that the substrate has a lithium ion conductivity of 7.0×10 ⁻⁵ S·cm ⁻¹ or more at 25° C. and a thickness of 300 μm or less. The thickness of the substrate is more preferably 200 μm or less, and further, although not limited thereto, the thickness of the substrate is preferably 0.5 μm or more since it is easy to reduce short circuits between the positive electrode and the negative electrode when the substrate is made into a solid electrolyte layer. Furthermore, it is very preferable that the maximum particle diameter of the particles in the outermost particle layer of the substrate is 30 μm or less and the average particle diameter of the particles is 15 μm or less, since the lithium ion conductivity is further increased. In addition, the reduction resistance of the substrate is preferably 1.5 V (vs. Li) or less at 25° C., similar to the powder, and more preferably 1.0 V (vs. Li) or less.

Here, in the present invention, the "outermost particle layer" of the substrate means a particle layer formed by particles exposed on the substrate surface in the glass ceramic of the present invention, which is the substrate. The same applies to forms other than substrates.
In the present invention, the "maximum particle size" and "average particle size" of a particle refer to the maximum value and the average value when all particle sizes (the length of the longest diagonal) are measured for particles whose entirety is completely contained within a 24 μm × 19 μm visual field by observation with a scanning electron microscope (SEM).

The glass ceramic of the present invention, which has the above-mentioned structure, is a composition in which the particles do not become coarse during firing, etc., but become dense, and the crystal structure is unlikely to change even when the glass ceramic is cooled from a high temperature. It can be easily manufactured industrially, and has both high lithium ion conductivity and high reduction resistance. This high reduction resistance results in a solid electrolyte with a wide potential window on the low potential side, so that a low potential electrode active material and a high potential electrode active material can be used in combination, resulting in a high voltage battery. Even when charging and discharging at a high voltage, decomposition due to reduction is unlikely to occur, and the lithium ion conductivity is stable for a long period of time. Therefore, it can be suitably used as a solid electrolyte for all-solid-state secondary batteries, seawater batteries, etc.

For example, when the glass ceramic of the present invention is used as a solid electrolyte of an all-solid-state secondary battery, it can be used in at least one selected from the solid electrolyte layer and the electrode layer (positive electrode layer, negative electrode layer). In particular, since the glass ceramic of the present invention has high resistance to reduction, it is suitable for use in the solid electrolyte layer or negative electrode layer. These will be described in detail below.

<Solid electrolyte layer>
A solid electrolyte layer for an all-solid-state secondary battery can be formed by mixing the glass ceramic of the present invention with an inorganic binder or the like as necessary and then sintering the mixture. In particular, from the viewpoint of further increasing lithium ion conductivity, the solid electrolyte layer preferably contains 80 mass % or more of the glass ceramic of the present invention, more preferably contains 90 mass % or more, and more preferably is made of the glass ceramic of the present invention. In the case of a plate-shaped solid electrolyte layer, its thickness is preferably 0.5 μm or more and 300 μm or less, similar to the above-mentioned substrate.

In addition, when an inorganic binder is used, it is preferable to use a lithium ion conductive inorganic binder, and examples of such inorganic binders include amorphous or polycrystalline _LiPO3 , _{70LiPO3-30Li3PO4} , _Li2O - _SiO2 , _Li2O - _SiO2 - _{P2O5 -B2O5 - BaO} , etc. In particular, one or more selected _{from Li2O} - _{P2O5 - based} glass, _Li2O - _P2O5 - _{M2O3 -} based glass ₍ including those in which P is replaced by Si, where M is Al or B), and _LiPO3 melted and then quenched to make it amorphous are preferable. In a particularly preferred embodiment, _LiPO3 is melted and then rapidly cooled to make it amorphous, which has a low glass transition temperature (Tg) of about 280° C. and is unlikely to crystallize, so that a solid electrolyte layer having high lithium ion conductivity and high reduction resistance can be formed by mixing this inorganic binder with the glass ceramic of the present invention and heating it to 600° C. The content of the inorganic binder contained in the solid electrolyte layer is preferably 20 mass % or less, more preferably 10 mass % or less, and even more preferably 5 mass % or less, based on the total mass of the solid electrolyte layer.

<Electrode layer>
The glass ceramic of the present invention is mixed with an electrode active material (positive electrode active material or negative electrode active material) and, if necessary, a conductive assistant, an inorganic binder, etc., and then sintered to form an electrode layer for an all-solid-state secondary battery. In particular, the electrode layer preferably contains 20% by mass or more of the glass ceramic of the present invention, more preferably 40% by mass or more, and even more preferably 50% by mass or more. This is because the lithium ion migration path is easily secured in the electrode layer, making it easier to improve the charge/discharge characteristics and battery capacity of the battery. On the other hand, in this electrode layer, the glass ceramic of the present invention is preferably 80% by mass or less, more preferably 70% by mass or less. This is because the filling amount of the electrode active material is secured, and the battery capacity can be increased.

Examples of the positive electrode active material include NASICON type _LiV2 ( _PO4 ) ₃ , olivine type _LixJyMtPO4 ( _wherein J is at least one selected from Al, Mg, and W, Mt is at least one selected from Ni, Co, Fe, and Mn, and x satisfies 0.9≦x≦1.5 and y is 0≦y≦0.2), _layered oxides, and spinel type oxides. In particular, _LiMtO2 and/or _LiMt2O4 (wherein Mt is at least one selected from Fe, Ni, Co, and Mn) are preferred. This is because the glass ceramic of the present invention and the positive electrode active material are easily bonded to each other, _and lithium ions are easily transferred between them, thereby further improving the charge and discharge characteristics of the all-solid-state secondary battery. Specific examples of the positive electrode active material include LiCoPO ₄ , LiCoO ₂ , and LiMn ₂ O _4. Furthermore, by adding Mg as a trace component, it is possible to suppress thermal decomposition of the positive electrode active material and improve the discharge capacity.

In addition, the negative electrode active material may be at least one selected from, for example, oxides containing NASICON type, olivine type, or spinel type crystals, rutile type oxides, anatase type oxides, or amorphous metal oxides, or metal alloys. In addition, low-potential negative electrode active materials such as Li ₄ Ti ₅ O ₁₂ and TiO ₂ can also be used. In particular, at least one selected from Li _1+x+z Al _x Ti _2-x Si _z P _3-z O ₁₂ (wherein x satisfies 0≦x≦0.8 and z satisfies 0≦z≦0.6), Li ₄ Ti ₅ O ₁₂ , and TiO ₂ is more preferable. This is because the glass ceramic of the present invention and the negative electrode active material are easily bonded, and lithium ions are easily transferred between them, thereby further improving the charge and discharge characteristics of the all-solid-state secondary battery. Specific examples of the _{negative electrode} active material include _Li2V2 ( _PO4 ) ₃ , _Li3Fe2 ( _PO4 ) ₃ , _LiFePO4 , _Li4Ti5O12 , _SiOx ( _0.25 ≦ _x ≦2 ₎ , and _Cu6Sn5 .

The content of these electrode active materials in the electrode layer is preferably 10% by mass or more and 50% by mass or less. In particular, by making this content 10% by mass or more, the battery capacity of the all-solid-state secondary battery can be further increased. And, more preferably, it is preferable to set the lower limit at 20% by mass. On the other hand, by making this content 50% by mass or less, it is easier to ensure the ionic conductivity of the electrode layer. And, more preferably, it is preferable to set the upper limit at 40% by mass, and even more preferably at 30% by mass.

Examples of conductive assistants used in the electrode layer include carbon compounds such as graphite, activated carbon, and carbon nanotubes; at least one metal selected from Ni, Fe, Mn, Co, Mo, Cr, Ag, and Cu; alloys of these metals; metals such as titanium, stainless steel, and aluminum; and precious metals such as platinum, gold, ruthenium, and rhodium. By using such highly electronically conductive materials as conductive assistants, the amount of current that can be conducted through the narrow electronic conduction path formed in the electrode layer is increased, thereby improving the charge and discharge characteristics of the all-solid-state secondary battery.

Taking into consideration the balance between the battery capacity and the electronic conductivity of the electrode layer, the content of this conductive assistant is preferably 1% by mass or more and 20% by mass or less, more preferably 2% by mass or more and 15% by mass or less, and most preferably 4% by mass or more and 10% by mass or less, of the total electrode material (i.e., positive electrode active material or negative electrode active material) contained in the electrode layer.

The inorganic binder used in the electrode layer can be the same as that used in the solid electrolyte layer described above. Furthermore, a current collector may be provided in at least one of the electrode layers, i.e., the positive electrode layer and the negative electrode layer. This is because it is easy to extract electricity through the current collector, which makes it easier to charge the all-solid-state secondary battery and discharge the all-solid-state secondary battery. The current collector may be a thin-film metal layer laminated or bonded to the positive electrode layer and/or the negative electrode layer, or may be a material obtained by laminating a metal layer or a precursor of a conductor on the raw material composition and then firing it. If the electronic conductivity of the electrode layer itself is high, this current collector may not be provided.

An all-solid-state secondary battery can be formed using the above-mentioned solid electrolyte layer and electrode layer, but one or two of the solid electrolyte layer, the positive electrode layer, and the negative electrode layer may be conventionally known, and such conventionally known layers may be combined with a solid electrolyte layer or an electrode layer containing the glass ceramics of the present invention. However, by making the solid electrolyte layer, the positive electrode layer, and the negative electrode layer all contain the glass ceramics of the present invention, an all-solid-state secondary battery with better charge/discharge characteristics can be formed.

The contents and compositions of the glass ceramics, electrode active material, and conductive assistant of the present invention can be determined by cutting out the solid electrolyte layer or electrode layer and using an energy loss analyzer or an X-ray analyzer mounted on a field emission transmission electron microscope (FE-TEM), or an X-ray analyzer mounted on a field emission scanning electron microscope (FE-SEM). When an X-ray analyzer is used, the Li ₂ O component cannot be directly analyzed, but the content of the Li ₂ O component can be estimated by calculating the charge from the other constituent components.

<Method for producing glass ceramics of the present invention>
Next, the method for producing the glass ceramics of the present invention will be described in detail.
The glass ceramics of the present invention have a major feature that they can be easily industrially produced using general methods for producing inorganic materials, such as firing or melting of inorganic materials, mixed firing, etc. Although not limited thereto, it is preferable to produce the glass ceramics of the present invention by a solid-phase method including a mixing step of mixing a raw material composition, and a firing step of firing the mixed raw material composition or forming the raw material composition into a desired shape and then firing it to generate a desired crystal phase.

For example, one example of a method for producing the glass ceramics of the present invention includes a raw glass preparation step of melting and vitrifying a lithium ion conductive raw material containing an inorganic substance containing at least Li and/or P to obtain a raw glass, a crushing step of crushing the raw glass and mixing it with an amorphous inorganic substance containing Zr to obtain a powdered amorphous precursor, and a firing step of firing the amorphous precursor at 1100 to 1300° C. to precipitate crystals, and a glass ceramics composition containing, in mole percent based on _oxide , 10 to 20% _Li2O component, 30 to 40% _P2O5 component, 40 to 50% _ZrO2 component, 0 to ₄ % _Y2O3 component _{, 0 to 3% Al2O3 component} , and GeO and a firing step for obtaining a lithium ion conductive glass-ceramic containing 0 to 2% of _{the two} components, containing a crystal phase having a rhombohedral NASICON structure, and having an a-axis lattice constant of 8.872 Å or more as identified by X-ray diffraction and Rietveld analysis. The firing step may be performed in two or more steps, but is preferably performed in one step from the viewpoint of production efficiency, etc.

In the above-mentioned example of the manufacturing method, a forming step is included in which the powdered amorphous precursor is formed into a substrate of the desired size and thickness prior to the firing step, and the amorphous precursor formed body obtained by this forming step is fired in the above-mentioned firing step to obtain a substrate-shaped lithium ion conductive glass ceramic (the glass ceramic of the present invention which is a substrate). Even when manufacturing the glass ceramic of the present invention in a form other than a substrate, the powdered amorphous precursor can be formed into the desired shape in a similar forming step.

In addition, when producing the glass ceramics of the present invention in a form other than powder, the method of firing the powdered amorphous precursor by the firing process described above and then molding the fired powder into the desired shape (a method in which the firing process is not performed after molding, and a method in which molding and re-firing after firing) cannot obtain the glass ceramics of the present invention having low grain boundary resistance, high lithium ion conductivity, and high reduction resistance. The same is true when mixing with a polymer or the like and molding into a sheet.

The embodiment described above is merely an example for facilitating understanding of the present invention, and does not limit the present invention. In other words, the components described above may be modified or improved without departing from the spirit of the present invention, and the present invention naturally includes equivalents thereof.

Below, we will explain the examples of the present invention, but the present invention is not limited to the following examples, and various modifications are possible within the technical concept of the present invention.

<Preparation of solid electrolyte (Comparative Example 1) by solid phase method using precursor not having amorphous composition>
Zirconia oxide ( _ZrO2 ), _lithium carbonate ( _Li2CO3 ), ammonium dihydrogen phosphate ( _NH4H2PO4 ), and yttrium oxide ( _Y2O3 ) were used as raw material powders, and mixed _to have a _{stoichiometric} ratio of _{Li1.15Y0.15Zr1.85P3O12} . The mixture was then placed in _{a 600} ml PP cup and mixed for 3 _hours in a pot mill using a _YTZ ball _with an outer diameter of 10 mm. The resulting mixed powder was then placed in a platinum pot and fired at 900°C for 3 hours to obtain a precursor. The heating rate was 300°C/h.
The obtained precursor was then pulverized using an alumina mortar and pestle to obtain a 106 μm mesh pass. It was then pulverized in a planetary port mill until the cumulative 90% particle size (D90) was 1 μm or less, and dried. The obtained dried powder was then placed in a platinum pot and fired in a Kanthal furnace at 1200° C. for 20 hours, and then cooled in the furnace to obtain a solid electrolyte powder of Comparative Example 1.

In addition, the above-mentioned dried powder before sintering was crushed in an agate mortar, and 1.5 g of a sample that had passed a 0.5 mm mesh was then molded into tablets using a mold with an outer diameter of 20 mm at a force of 20 kN. The resulting molded body was placed on a platinum plate and fired in a Kanthal furnace at 1200°C for 20 hours and then cooled in the furnace, thereby obtaining solid electrolyte pellets of Comparative Example 1.

The above-mentioned cumulative 90% particle size (D90) means the cumulative 90% by volume particle size from the smallest particle size in the particle size distribution, and was measured and calculated using a particle size distribution measuring device (Mastersizer 3000, manufactured by Spectris) using the laser diffraction/scattering method in accordance with JIS R 1629 "Method for measuring particle size distribution of fine ceramic raw materials by laser diffraction/scattering method."

<Preparation of solid electrolytes (Examples 1 to 10) by solid phase method using precursors having amorphous composition>
The amorphous zirconium-containing inorganic raw material and a raw material glass obtained by vitrifying the other raw materials were pulverized, mixed, and then dried to obtain a precursor having an amorphous composition, which was then fired or formed into tablets and fired to obtain the solid electrolytes of Examples 1 to 10, which are glass ceramics of the present invention. In the drawings described later, Example 2 of these Examples may be indicated as "LYZSP12" or "LYZSP".
The method will be described in detail below, step by step.

(Preparation of raw glass)
First, lithium metaphosphate (LiPO ₃ ), trilithium phosphate (Li ₃ PO ₄ ), yttrium oxide (Y ₂ O ₃ ), or silicon dioxide (SiO ₂ ) was prepared in the amounts shown in Table 1 or 2 below. Each prepared sample was placed in a platinum pot, melted with thorough stirring at 1100° C. or higher, vitrified, and cast onto a metal casting plate. The yield of the recovered raw glass, including the raw glass adhering to the platinum pot, was 99% or more by weight in all of Examples 1 to 10.

(Preparation of glass-ceramic solid electrolyte)
After each of the raw glasses was crushed to pass a 106 μm mesh, amorphous zirconium phosphate ((ZrO) ₂ (HPO ₄ ) ₂ ) was added in the amounts shown in Table 1 below for Examples 1 to 9, and yttrium oxide (Y ₂ O ₃ ) and amorphous zirconium phosphate ((ZrO) ₂ (HPO ₄ ) ₂ ) were added in the amounts shown in Table 2 below for Example 10, and crushed by a planetary ball mill using 1-propanol as a dispersion medium. The crushing medium was YTZ beads (manufactured by Nikkato Corporation) with an outer diameter of 2 mm. The crushed slurry was dried, and the dried powder was placed in a platinum pot as an amorphous precursor, and fired in a Kanthal furnace at 1200° C. for 20 hours and cooled in the furnace to obtain solid electrolyte powders of Examples 1 to 10.
In addition, each of the above-mentioned dried powders before firing was crushed in an agate mortar, and 1.5 g of a sample that had passed a 0.5 mm mesh was then molded into a tablet using a mold having an outer diameter of 20 mm with a force of 20 kN. The obtained molded body was placed on a platinum plate and fired in a Kanthal furnace at 1200° C. for 20 hours and then cooled in the furnace, thereby obtaining solid electrolyte pellets of Examples 1 to 10.

<Density and lithium ion conductivity measurements>
The surfaces of the solid electrolyte pellets of Comparative Example 1, Example 1, and Example 2 were polished with #800 and #2000 waterproof abrasive paper and 1-propanol, and then the diameter, thickness, and weight were measured with a caliper, a micrometer, and an electronic balance, and the density was calculated. The results are shown in Table 3 below.

Furthermore, gold electrodes were formed as blocking electrodes on both sides of each solid electrolyte pellet using a magnetron sputtering device (SC-701HMC, manufactured by Sanyu Electronics Co., Ltd.). Then, impedance measurement was performed at 25°C under the conditions of a frequency of 0.1 Hz to 7 MHz, an amplitude voltage of 10 mV, and an open circuit voltage using an electrochemical evaluation device (SP300, manufactured by Biologic Co., Ltd.), and the lithium ion conductivity was calculated. The results are also shown in Table 3 below. For reference, a Cole-Cole plot obtained by the impedance measurement in Example 2 is shown in Figure 1. The lithium ion conductivity was calculated based on the resistance at the bending point on the low frequency side (the part indicated by the arrow in Figure 1 in Example 2), the thickness, and the electrode area.
It was shown that the solid electrolytes of Examples 1 and 2 had lithium ion conductivities higher than that of Comparative Example 1 by one order of magnitude or more.

<Powder X-ray diffraction and Rietveld analysis>
In order to confirm the crystal structure of the solid electrolyte, powder X-ray diffraction and Rietveld analysis were performed. First, the solid electrolyte powders of Comparative Example 1, Example 1, and Example 2 were crushed in an alumina mortar, and then 1 g of the crushed sample and 0.1 g of zinc oxide (ZnO, manufactured by Kojundo Chemical Laboratory Co., Ltd.) as a reference were added and mixed for 5 minutes using an agate mortar and pestle, and powder X-ray diffraction measurement was performed. The powder X-ray diffraction measurement was performed using a D8 DISCOVER manufactured by Bruker Co., Ltd., and the weight measurement was performed using an electronic balance capable of weighing to the nearest 0.1 mg, and the weight was measured while being neutralized using an ionizer so that there was no weight change before and after the measurement due to static electricity, etc. Calibration of ZnO was performed using α-Al ₂ O ₃ (Standard Reference Material 674, manufactured by U.S. Department of Commercial National Bureau of Standards). Calibration of ZnO was performed by mixing 0.181 g of ZnO and 0.122 g of the above-mentioned α-Al ₂ O ₃ in an agate mortar, and measuring the powder X-ray diffraction of the mixture in the same manner as described above.

In addition, the Rietveld analysis was performed by analyzing the above-mentioned powder X-ray diffraction measurement results using the Rietveld analysis software "Z-Rietveld code" with the profile function "Split Pseudo Voigt function". The initial parameter values of the crystal structure were 97658 for the α phase (LiZr ₂ (PO ₄ ) ₃ α), 26170 for ZnO, 89456 for the α' phase (LiZr ₂ (PO ₄ ) ₃ α'), 201131 for YPO ₄ , and 80046 for ZrO ₂ in terms of CSD number. The space group was fixed as it was initially for all phases, and all values allowed by the space group for the lattice constant were variable for all phases.

The results of this Rietveld analysis are shown in Table 4 below. The structural parameters of the α phase of Comparative Example 1 obtained by this Rietveld analysis are shown in Table 5 below, the structural parameters of the α phase of Example 1 in Table 6 below, and the structural parameters of the α phase of Example 2 in Table 7 below. Phases other than the crystalline phase (unspecified phase, mainly amorphous phase) were calculated using the following formula.

(Calculation formula for phases other than crystalline phase)
Mass ratio of phases other than the crystalline phase (normalized by ZnO)=(mass ratio of evaluation sample with mixed ZnO set to 1−sum of mass ratios of each crystalline phase other than ZnO obtained by Rietveld analysis with ZnO set to 1)÷mass ratio of evaluation sample with mixed ZnO set to 1 (formula for calculating mass ratio of phases other than the crystalline phase excluding ZnO)
Mass ratio of phases other than the crystalline phase (normalized by ZnO) / Sum of mass ratios of each crystalline phase other than ZnO × 100%
The correction factor for ZnO, calibrated by Rietveld refinement, was 0.879.

As can be seen from these results, the solid electrolyte powders of Comparative Example 1, Example 1, and Example 2 all contain a rhombohedral NASICON structure _LiZr2 ( _PO4 ) ₃ α-phase crystal phase, but Comparative Example 1 has almost no phases other than the crystal phase, whereas Examples 1 and 2 have phases other than the crystal phase mainly composed of amorphous phases of more than 3 mass% (i.e., the sum of the masses of the crystal components is less than 97 wt% of the total mass). In addition, the a-axis lattice constants of these α-phases are less than 8.872 Å in Comparative Example 1, whereas Examples 1 and 2 exceed 8.882 Å.

<Secondary electron image observation>
Since the state of the particles has a large effect on the lithium ion conductivity, secondary electron images of the outermost particle layer were also observed for the solid electrolyte pellets of Comparative Example 1, Example 1, and Example 2. The secondary electron images were observed using an S-3000N manufactured by Hitachi High-Technologies Corporation. The results are shown in FIG. 2 for the secondary electron image of Comparative Example 1, FIG. 3 for the secondary electron image of Example 1, and FIG. 4 for the secondary electron image of Example 2. The observation conditions were as follows: for Comparative Example 1, the working distance (WD; working distance (distance from the center of the objective lens to the sample surface)) was 16.5 mm, the acceleration voltage was 15 kV, and the magnification was 500 times. For Example 1, the WD was 15.9 mm, the acceleration voltage was 15 kV, and the magnification was 500 times. For Example 2, the WD was 11.5 mm, the acceleration voltage was 15 kV, and the magnification was 500 times.
Even under the same firing conditions, the solid electrolyte of Comparative Example 1 obtained from a precursor having no amorphous composition showed significant grain growth (FIG. 2). On the other hand, in the solid electrolytes of Examples 1 and 2 obtained from precursors having an amorphous composition, the formation of clear necks was confirmed, and the grain boundaries were densified (FIGS. 3 and 4).

<Preservation test>
Solid electrolytes are mainly used for batteries, and in battery applications, it is necessary for the lithium ion conductivity to be stable for a long period of time, so a storage test was performed. As an evaluation method, the solid electrolyte pellets of Comparative Example 1, Example 1, and Example 2 were stored in the atmosphere at 25°C, and powder X-ray diffraction and Rietveld analysis, as well as density and ion conductivity measurements were performed in the same manner as described above. The storage period was one year.

Table 8 below shows the results of the Rietveld analysis before the start of the storage test (initial product) and after one year of storage (one year later), and Table 9 below shows the density and lithium ion conductivity measurement results. In the solid electrolyte of Example 1 not containing a Si component, the α phase decreased from 96.6 mass% to 28.8 mass% after one year of storage, and decomposed into α' phase, YPO ₄ , ZrO ₂ , etc. On the other hand, the decomposition (mass ratio decrease) of the α phase in the solid electrolyte of Example 2 containing a Si component was suppressed to less than 10 mass%. And, the lithium ion conductivity of the solid electrolyte of Example 1 not containing a Si component decreased by about 46% after one year of storage, while the lithium ion conductivity of the solid electrolyte of Example 2 containing a Si component decreased by about 28%.

The results of this storage test are compared with _the literature values of _similar products ( _{Li1.15Y0.15Zr1.85P3O12} ( _JCPDS 01-083-4639), _{Li1.2Ca0.1Zr1.9P3O12} ( _RSC Advances ₂₀₁₁ (1 ₎ 1728-1731) _{, LiZr2P3O12} rhombohedral crystal system (JCPDS 01-084-0998)) in terms of lattice constants ( _a -axis (a/Å), c-axis (c/Å)) and lithium ion conductivity, as shown in Table _10. FIG. ₅ shows plots of the initial values and the lithium ion conductivity and the a-axis lattice constant after one year for Examples 1 and 2. The a-axis lattice constants of the α-phase in Examples 1 and 2, which were made from a precursor having an amorphous composition, are larger than the a-axis lattice constants in Comparative Example 1, which was made from a precursor not having an amorphous composition, and the literature values of similar products, and in particular, Example 2, which contains a Si component, was the largest. From the relationship between lithium ion conductivity and the a-axis lattice constant, it can be confirmed that the lithium ion conductivity tends to increase as the a-axis lattice constant increases. This is presumably because lithium ions can move more freely within the lattice of the crystalline phase.
In other words, it was confirmed that the solid electrolyte of Example 2 containing a Si component is particularly excellent in stability at room temperature (25° C.), and that by replacing a part of P with Si, the a-axis direction of the lattice is further extended, thereby increasing the lithium ion conductivity, and further increasing the stability of the crystal structure over time and the stability of the lithium ion conductivity.

<Confirmation of oxidation-reduction potential by cyclic voltammetry>
The reduction resistance of the solid electrolytes of Examples 1 and 2 was evaluated by cyclic voltammetry (CV) using lithium as a counter electrode.

A Li metal foil (electrode) and a polymer electrolyte were attached to the copper foil as a current collector, and the solid electrolyte pellet of Example 1 or Example 2, on which an Au electrode had been sputtered, was placed on the working electrode side, and current was collected with aluminum foil. This evaluation sample was vacuum-packed in an aluminum laminate pack, and only the copper foil and aluminum foil of the current collector could be taken out with a tab film. As a pretreatment of the sealed evaluation sample, it was kept at 60°C for 2 hours or more to promote bonding between the polymer electrolyte and the solid electrolyte, and then evaluation was performed.
The measurement temperature was 25° C., the starting voltage was the open circuit voltage, and the scanning rate was 0.2 mV/s, scanning from 5 V to 0.2 V. The obtained cyclic voltammogram (current-potential curve: CV curve) is shown in Fig. 6. It was confirmed that the solid electrolyte of Example 2 containing a Si component had little change in current and was more excellent in reduction resistance.

<Confirmation of the effective valence of Si by X-ray absorption fine structure analysis (XAFS)>
From the above results, it was confirmed that the solid electrolyte of Example 2 has high lithium ion conductivity and high reduction resistance. However, since the effect of the Si component and the state of Si in the solid electrolyte of Example 2 were unknown, the valence of Si was confirmed by X-ray absorption fine structure analysis (XAFS) for the solid electrolyte pellet of Example 2. The X-ray absorption fine structure analysis was performed at BL6N1 of the Aichi Synchrotron Light Center. The results are shown in FIG. 7.

The results in Figure 7 show good agreement with the measured and calculated XAFS spectrum for SiO ₆ coordination (Photon Factory Activity Report 2012 # 30 (2013) B). This suggests that Si is highly likely to be in SiO ₆ coordination in the solid electrolyte, and is highly likely to be present in the ZrO ₆ site in the crystal or in the amorphous part of the solid electrolyte.

The arrows in Fig. 7 indicate the areas where the difference from _SiO2 is significant. In particular, from the new peak that appears at about 1845 eV indicated by the arrow on the left, it can be interpreted that the solid electrolyte of Example 2 is contaminated with Si having a valence lower than 4. Since the difference in absorption edge energy between Si equivalent to 0 valence and _SiO2 (equivalent to 4 valence) is 9 eV, it was confirmed from the energy difference of the above peak (about 3 eV) that Si having a valence of about 3.7 is included.

<Confirmation of the coordination number of O to Si by extended X-ray absorption fine structure analysis (EXAFS)>
Furthermore, the coordination number of O to Si in the solid electrolyte of Example 2 was also confirmed by extended X-ray absorption fine structure analysis (EXAFS). For the extended X-ray absorption fine structure analysis, synchrotron radiation data measured at BL6N1 of the Aichi Synchrotron Center, which was used in the above-mentioned XAFS measurement, was used. The analysis conditions were as follows.

The K-edge of Si is E = 1847 eV, while the LIII, LII, and LI-edges of Y are E = 2080 eV, 2156 eV, and 2373 eV, respectively, and the LIII, LII, and LI-edges of Zr are E = 2222 eV, 2307 eV, and 2532 eV, respectively, and are in the K-edge EXAFS region of Si. The data in Figure 8 shows the contamination of the LIII and LII-edges of Y. Therefore, in the EXAFS analysis of the solid electrolyte of Example 2, it is necessary to limit the E range (wave number k range) used in the analysis to avoid these contaminations. Therefore, in order to improve the accuracy of the analysis as much as possible even within this limit, the following measures were taken.

In order to reproduce the experimental data with as few variables as possible and to determine the nearest-neighbor O coordination number, the surrounding atoms were limited to the nearest-neighbor O. As a result, the fitting variables were the scale factor S ₀ ² O coordination number (No), the Si-O distance (r _Si-O ), the absorption edge (E ₀ ), and the Debye-Waller factor (σ ₀ ² ).
Then, the fitting variables were determined in sequence according to the following flow, with the restricted E range corresponding to 1897 to 2061 eV kmax = 7.4, the wavenumber weighting order: second order, and the fitting space: wavenumber space.
(1) The reference SiO ₂ is analyzed over a wide E range to determine σ ₀ ² .
(2) The reference SiO ₂ is analyzed within the restricted E range (σ ₀ ² is fixed to the value in (1)), and S ₀ ² is determined from this data which can be fixed as No=4.
(3) Example 2 (LYZSP12) is analyzed within the restricted E range (σ ₀ ² is fixed to the value in (1) and S ₀ ² is fixed to the value in (2)), and the No of the target crystal phase is evaluated.

The results of this analysis are shown in Table 11 below and in Figure 9. Figure 9 shows the "radial structure function in real space" extracted by Fourier transform. Here, the R range and k (or q) range used in the extraction are illustrated as window functions. Based on the agreement between the function extracted from this data and the function of the fitting result, and the R factor in Table 11, it was determined that this analysis was performed appropriately.
The coordination number of O to Si in the solid electrolyte of Example 2 was estimated to be 6, not 4 (Table 11). This result suggests that the inclusion of a Si component in the solid electrolyte strengthens the covalent bonds in the crystalline phase (particularly in the α phase), and Si is placed at the Zr site to energetically stabilize the lattice, or that there is a high possibility that the amorphous phase is stabilized by 6-coordinated Si.

<Confirmation of the effect of the ratio of Y and Si on lithium ion conductivity and density>
_The influence of the ratio of Y (the value of x in the formula) in _{Li1.05+xYxZr2 - xSi0.05P2.95O12 in} the solid electrolyte pellets of Examples 3 to ₆ on the lithium ion conductivity, density, and a-axis lattice constant of the α phase was confirmed with reference to Example 2. The results are shown in Table 12 below and FIG. 10.
In addition, the influence of the Si ratio (the value of _y in the formula) of _{Li1.15+yY0.15Zr1.85SiyP3 - yO12 in} the solid electrolyte pellets of Examples 7 to ₉ on the lithium ion conductivity, density, and a-axis lattice constant of the α-phase was confirmed with reference to Examples 1 and 2. The results are shown in Table 13 and FIG.
Furthermore, _the influence of the a-axis lattice constant of the α- _phase , which _is changed by varying the Y ratio (the value of x in the formula) of Li1.05 _{+xYxZr2 - xSi0.05P2.95O12 and} the Si ratio (the _value of y in the formula) of Li1.15 _{+yY0.15Zr1.85SiyP3 - yO12 ,} on the lithium ion conductivity was confirmed. The results are shown in FIG. 12.
Furthermore, the Si ratio (the value of y in the formula) of Li1 _{+xYxZr2 -xSiyP3 - yO12 was} set to 0, and the influence _of the Y ratio (the value of x in the formula) _of Li1 _{+xYxZr2 - xP3O12 on} the lithium ion conductivity, density, and a-axis lattice constant of the α-phase was examined. The results are shown in Table 14.

The lithium ion conductivity, density, and a-axis lattice constant of the α-phase of each solid electrolyte pellet were measured in the same manner as described above. The lithium ion conductivity and density were measured twice for each example (Tables 12 to 14, Figures 10 to 12).

As a result, it was shown that when the _Y ratio in Li1.05 _{+xYxZr2 - xSi0.05P2.95O12 is} 0.10 to _0.17 , and when the Si ratio in Li1.15 _{+yY0.15Zr1.85SiyP3 - yO12 is 0.00} to _0.07 , the lithium ion conductivity is further increased.
_It was also confirmed that the a-axis lattice constant of the α-phase varies from 8.880 Å _{to 8.8892 Å} by varying the ratio of Y in Li1.05 _{+xYxZr2 - xSi0.05P2.95O12 or} the ratio of Si in _{Li1.15 +yY0.15Zr1.85SiyP3 - yO12 ,} and that the value of the a-axis lattice constant of the α-phase correlates with the value of the lithium ion conductivity. In other words, it was confirmed that the lithium ion conductivity increases as the a-axis lattice constant of the α-phase increases from 8.880 Å, that the lithium ion conductivity is highest when the a-axis lattice constant of the α-phase is in the vicinity of 8.885 Å, and that the lithium ion conductivity tends to decrease when the a-axis lattice constant of the α-phase exceeds 8.892 Å.
Furthermore, even when the Y ratio in Li1 _{+x+yYxZr2 -xSiyP3 - yO12 exceeds} 0.2, i.e., 0.25, it was confirmed that, if the Si ratio is 0, a high lithium ion conductivity is relatively maintained compared to the case where _the Si ratio is 0.05. The a-axis lattice constant of the α phase in Example 10 is 8.888 Å, which is considered to be influenced by the fact that this is equal to or less than 8.892 Å, which is the threshold value of the lattice constant of the a-axis of the α phase.

<Preparation of sintered body by sheet molding method>
For the solid electrolyte of Example 2, which was confirmed to have high lithium ion conductivity and stable performance in the form of pellets or powder, a sheet-like sintered body was produced by green sheet molding.
Specifically, the dried powder of the amorphous precursor having the composition of Example 2 was pulverized to 1 μm or less at D90 to obtain a powder, to which 1-propanol as a solvent, a binder, and a dispersant were added, and the powder was molded into a sheet having a thickness of 20 μm. 12 sheets were stacked, vacuum packed, and then subjected to hot water isostatic pressing and sintered at 1200° C. The obtained sheet-like sintered body (sheet-like substrate) was designated as Example 11.

<Measurement of density and lithium ion conductivity of sheet substrate>
The diameter, thickness, and weight of the sheet substrate of Example 11 were measured using a caliper, a micrometer, and an electronic balance, respectively, and the density was calculated. The density was 2.6 to 2.85 g/cm ^3. Furthermore, gold electrodes were formed as blocking electrodes on both sides of the sheet substrate of Example 11 using a magnetron sputtering device (manufactured by Sanyu Electronics Co., Ltd., SC-701HMC). Then, this was placed in a thermostatic bath at 25°C, and impedance measurement was performed using an electrochemical evaluation device (manufactured by Biologic Co., Ltd., SP300) at a frequency of 7 MHz to 0.1 Hz and an amplitude voltage of 10 mV, and the lithium ion conductivity was measured. The Cole-Cole Plot is shown in FIG. 13. The lithium ion conductivity at 25° C. was 8.5 to 9.3×10 ⁻⁵ S·cm ⁻¹ , which was higher than the lithium ion conductivity (literature value) of a solid electrolyte (Li _1.15 Y _0.15 Zr _1.85 P ₃ O ₁₂ ) synthesized by SPS.

<Secondary electron image observation results>
The sheet substrate of Example 11 was observed using the same device as above, with a WD of 15.0 mm and an acceleration voltage of 15.1 kV. The magnification was set to 5000 times, and secondary electron images of the outermost particle layer were observed. The results are shown in Figures 14 and 15. From these results, it was confirmed that the grain boundaries were present, but that the grain interfaces were densely bonded.
In addition, for the particles present in the outermost particle layer, 27 particles that were completely contained within a 24 μm × 19 μm field of view were counted. The results are shown in Table 15 below. The particle size was taken as the length of the longest diagonal line (maximum value of the diagonal line) (FIG. 15). From these results, it was confirmed that sintering was possible with small particle sizes, with a maximum particle size of 10.2 μm, a minimum particle size of 0.7 μm, and an average particle size of 3.7 μm.

This application claims priority to Japanese Patent Application No. 2020-076009, filed on April 22, 2020, the disclosure of which is incorporated herein in its entirety.

Claims (9)

In mole percent based on oxide,
Li ₂ O component: 10 to 20%
_P2O5 component _30-40 %,
_ZrO2 component: 40 to 50%
_Y2O3 component 0 to ₄ %,
_Al2O3 component is 0 to ₃ %,
_GeO2 content: 0-2%
Contains
Contains a rhombohedral NASICON structure crystal phase,
the crystalline phase comprises a _LiZr2 (PO4 ₎ 3 _α -phase and/or a LiZr2 ₍ PO4 ₎ 3 _α' -phase;
The a-axis lattice constant of the crystalline phase identified by X-ray diffraction and Rietveld analysis is 8.872 Å or more.
Lithium-ion conductive glass-ceramics. The lithium ion conductive glass ceramic according to claim 1, in which the sum of the masses of the crystalline components identified by X-ray diffraction and Rietveld analysis is less than 97 mass% of the total mass. 3. The lithium ion conductive glass-ceramics according to claim 1 , wherein the mass of the _LiZr2 ( _PO4 ) ₃ α-phase identified by X-ray diffraction and Rietveld analysis is 80 mass% or more based on the total mass of the glass-ceramics. In mole percent based on oxide,
_SiO2 component: 0.1 to 5%
The lithium ion conductive glass ceramic according to any one of claims 1 to 3 , comprising: The lithium ion conductive glass ceramic according to claim 4 , wherein the valence of Si contained in the entire lithium ion conductive glass ceramic is 3.5 or more and less than 3.9. 6. The lithium ion conductive glass ceramic according to claim 4, wherein the coordination number of O to Si contained in the lithium ion conductive glass ceramic is 5 or more and 7 or less in the entire lithium ion conductive glass ceramic. 7. The lithium ion conductive glass ceramic according to claim 1 , which is a powder having a lithium ion conductivity at 25° C. of 1.0×10 ⁻⁵ S·cm ⁻¹ or more. 7. The lithium ion conductive glass ceramic according to claim 1 , which is a substrate having a lithium ion conductivity at 25° C. of 7.0×10 ⁻⁵ S·cm ⁻¹ or more and a thickness of 300 μm or less. 9. The lithium ion conductive glass ceramic according to claim 8 , wherein the maximum particle size of the particles in the outermost particle layer of the substrate is 30 μm or less, and the average particle size of the particles is 15 μm or less.

Images