JP7677296B2 - Inspection method and inspection system for all-solid-state batteries - Google Patents

Description

This disclosure relates to an inspection method and inspection system for solid-state batteries.

A method for inspecting an all-solid-state lithium battery (hereinafter referred to as an "all-solid-state battery") using a lithium ion conductive solid electrolyte has been proposed. For example, JP 2017-27770 A (Patent Document 1) discloses a method for inspecting the presence or absence of a micro-short circuit in a cell. This inspection method includes an external short-circuiting process for externally short-circuiting the positive and negative electrodes of the cell, a voltage measurement process for measuring the open circuit voltage between the positive and negative electrodes of the cell a certain time after the external short-circuiting process ends, and a determination process for determining that the cell is unusable if the open circuit voltage measured by the voltage measurement process is less than a threshold value.

JP 2017-27770 A JP 2014-134395 A International Publication No. 2019/175707 JP 2020-38836 A

In general, there is a constant demand for more efficient inspection of various types of batteries, including solid-state batteries. Improving the efficiency of inspection includes shortening inspection time, saving on inspection costs, and reducing the number of inspection steps.

The method described in Patent Document 1 requires assembling the cells for the external short-circuit process, and waiting for a certain period of time to pass (so-called aging) for the voltage measurement process. These processes leave room for improvement in terms of making the test more efficient.

This disclosure has been made to solve the above problems, and one of the objectives of this disclosure is to improve the efficiency of testing all-solid-state batteries.

(1) In the inspection method for an all-solid-state battery according to the first aspect of the present disclosure, the all-solid-state battery includes a plurality of coating layers. Each of the plurality of coating layers is formed by applying a slurry prepared through a kneading process of a solid electrolyte and a solvent to a current collector or a substrate. The plurality of coating layers includes a positive electrode layer formed by applying a positive electrode slurry, a negative electrode layer formed by applying a negative electrode slurry, and a separator layer formed by applying a separator slurry. The inspection method includes a step of acquiring a measurement result of the solid content rate of the corresponding slurry for at least one of the plurality of coating layers, and a step of determining that at least one coating layer is not suitable for use if the acquired solid content rate is outside a reference range determined based on the correspondence between the solid content rate of the corresponding slurry and the self-discharge amount of the all-solid-state battery.

(2) At least one coating layer includes a negative electrode layer. The acquiring step is a step of acquiring a measurement result of the solid content rate in the first stirring step of the negative electrode slurry.

(3) The reference range includes the range in which the solid content of the negative electrode slurry is greater than 43.30% and less than 43.65%.

(4) Using a trained model based on supervised learning, manufacturing parameters of the slurry that contribute to the self-discharge amount of the all-solid-state battery are extracted. The trained model has a decision tree. The decision tree uses the self-discharge amount as the objective variable and includes the solid content rate in the first stirring process of the negative electrode slurry as the explanatory variable.

(5) The inspection system for an all-solid-state battery according to the second aspect of the present disclosure includes one or more processors and a memory. The memory causes the one or more processors to execute the inspection method according to any one of claims 1 to 4 by executing a program by the one or more processors.

In the above methods (1) to (4) and configuration (5), the suitability of the coating layer for use is determined based on the solid content rate of the slurry. A coating layer determined to be unsuitable for use is not used in assembling an all-solid-state battery (cell). Therefore, it is possible to determine the quality of the all-solid-state battery before it is assembled. Therefore, the above methods and configurations can make the inspection of all-solid-state batteries more efficient.

This disclosure makes it possible to improve the efficiency of testing all-solid-state batteries.

FIG. 1 is a diagram illustrating an example of a configuration of an inspection system for an all-solid-state battery according to an embodiment of the present disclosure. FIG. 1 is a diagram showing an all-solid-state battery according to an embodiment of the present invention. 1 is a flowchart showing a procedure for preparing a negative electrode slurry in a comparative example. FIG. 11 is a diagram summarizing the contribution of manufacturing parameters of a negative electrode slurry to the self-discharge amount of a cell, obtained by machine learning. FIG. 13 is a graph showing the relationship between the solid content rate and the self-discharge amount in the first stirring step of the negative electrode slurry. FIG. 13 is a graph showing the relationship between the viscosity and the self-discharge amount in the second dispersion step of the negative electrode slurry. FIG. 13 is a diagram showing the relationship between particle size and self-discharge amount in the second stirring step of the negative electrode slurry. 3 is a flowchart showing a procedure for preparing a negative electrode slurry in the embodiment.

<Terminology>
In this specification, elements expressed in the singular form include the plural form unless otherwise specified. For example, a "particle" can mean not only "one particle" but also "an aggregate of particles (powder, powder, particle group)."

When a compound is expressed by a stoichiometric composition formula, the stoichiometric composition formula is merely a representative example of the compound. The compound may have a non-stoichiometric composition. For example, when lithium cobalt oxide is expressed as "LiCoO ₂ ", unless otherwise specified, lithium cobalt oxide is not limited to the composition ratio of "Li/Co/O=1/1/2" and may contain Li, Co and O in any composition ratio. Furthermore, doping, substitution, etc. with trace elements may also be allowed.

The following describes in detail the embodiments of the present disclosure with reference to the drawings. Note that the same or corresponding parts in the drawings are given the same reference numerals and their description will not be repeated.

[Embodiment]
<System Configuration>
1 is a diagram showing an example of the configuration of an inspection system for an all-solid-state battery according to an embodiment of the present disclosure. The inspection system 5 includes a server 51, an input device 52, a display 53, and a communication device 54. The server 51 includes a processor 511, a memory 512, a storage 513, and a network interface 514. The components of the inspection system 5 are connected to each other via a communication bus.

The processor 511 is, for example, a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit). The processor 511 reads a system program 61 and a control program 62 (described later), expands them in the memory 512, and executes them to realize various processes.

The storage 513 is a rewritable non-volatile memory such as a hard disk drive (HDD), a solid state drive (SSD), or a flash memory. The storage 513 stores a system program 61 including an operating system (OS), a control program 62 including computer-readable code required for control calculations, battery management data 63 including various manufacturing parameters (described below) for managing the solid-state battery, and a trained model 64 (described below).

The network interface 514 controls data communication between the server 51 and other external devices via the communication device 54.

The input device 52 is a keyboard, mouse, etc., and accepts input operations from an operator (measurer) who performs various measurements described below. The display 53 displays various information to the operator.

Note that while FIG. 2 shows an example in which the server 51 includes one processor 511, the server 51 may include multiple processors. That is, the server 51 includes one or more processors. The same applies to the memory 512 and the storage 513.

In this specification, the term "processor" is not limited to a processor in the narrow sense that executes processing using a stored program method, but may also include hardwired circuits such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array). Therefore, the term "processor" can also be interpreted as a processing circuitry whose processing is predefined by computer-readable code and/or hardwired circuits.

<All-solid battery>
2 is a diagram showing an all-solid-state battery in embodiment 2. A cell 100 of the all-solid-state battery includes a positive electrode 1, a negative electrode 2, and a separator layer 3. The cell 100 may include an exterior body (not shown) for housing an electricity storage element including the positive electrode 1, the negative electrode 2, and the separator layer 3. The exterior body is, for example, a pouch made of a metal foil laminate film. The positive electrode 1 and the negative electrode 2 correspond to the "positive electrode layer" and the "negative electrode layer" according to the present disclosure, respectively.

≪Positive electrode≫
The positive electrode 1 includes a positive electrode active material layer 11 and a positive electrode current collector 12. The positive electrode active material layer 11 is formed by applying a positive electrode slurry (a slurry prepared by kneading the material of the positive electrode active material layer 11 and a solvent) to the surface of the positive electrode current collector 12 and drying it. The positive electrode active material layer 11 is in close contact with the separator layer 3. The positive electrode active material layer 11 may have a thickness of, for example, 10 to 200 μm.

The positive electrode active material layer 11 includes positive electrode active material particles and a sulfide solid electrolyte. The positive electrode active material particles may include at least one selected from the group consisting of LiCoO ₂ , LiNiO ₂ , LiMnO ₂ , LiMn ₂ O ₄ , Li(NiCoMn)O ₂ , Li(NiCoAl)O ₂ , and LiFePO _4. The sulfide solid electrolyte includes S. The sulfide solid electrolyte may include Li, P, and S. The sulfide solid electrolyte may further include O, Si, or the like. The sulfide solid electrolyte may further include a halogen, or the like. The sulfide solid electrolyte may further include I, Br, or the like. The sulfide solid electrolyte may be, for example, a glass ceramic type or an argyrodite type. The sulfide solid electrolyte may include, for example, at least one selected from the group consisting _of LiI-LiBr-Li ₃ PS ₄ , Li ₂ S-SiS ₂ , LiI- _{Li 2 S} - _{SiS 2} , LiI-Li ₂ S-P _{2 S 5} , LiI-Li ₂ O-Li 2 S-P ₂ S ₅ , LiI- _{Li 2} S-P 2 O ₅ , LiI- _{Li 3} PO ₄ -P ₂ S ₅ , Li 2 S-P 2 S 5 , and Li ₃ PS ₄ .

The positive electrode active material layer 11 may further contain, for example, a conductive material. The conductive material may form an electronic conduction path within the positive electrode active material layer 11. The amount of the conductive material may be, for example, 0.1 to 10 parts by mass per 100 parts by mass of the coated active material. The conductive material may contain any component. The conductive material may contain, for example, at least one selected from the group consisting of carbon black, vapor grown carbon fiber (VGCF), carbon nanotubes (CNT), and graphene flakes.

The positive electrode active material layer 11 may further contain, for example, a binder. The amount of the binder may be, for example, 0.1 to 10 parts by mass per 100 parts by mass of the coated active material. The binder may contain any component. The binder may contain, for example, at least one selected from the group consisting of polyvinylidene fluoride (PVdF), vinylidene fluoride-hexafluoropropylene copolymer (PVdF-HFP), styrene butadiene rubber (SBR), and polytetrafluoroethylene (PTFE).

The positive electrode collector 12 may include, for example, an Al foil. The positive electrode collector 12 may have a thickness of, for example, 5 to 50 μm.

≪Negative electrode≫
The negative electrode 2 includes a negative electrode active material layer 21 and a negative electrode current collector 22. The negative electrode active material layer 21 is formed by applying a negative electrode slurry (a slurry prepared by kneading the material of the negative electrode active material layer 21 and a solvent) to the surface of the negative electrode current collector 22 and drying the slurry. The negative electrode active material layer 21 is in close contact with the separator layer 3. The negative electrode active material layer 21 may have a thickness of, for example, 10 to 200 μm.

The negative electrode active material layer 21 includes negative electrode active material particles and a sulfide solid electrolyte. The negative electrode active material particles may include at least one selected from _the group consisting of graphite, Si, SiOx ( ₀ <x<2), and _Li4Ti5O12 . The negative electrode active material layer 21 may further include a conductive material and a binder. The sulfide solid electrolyte between the positive electrode active material layer 11 and the negative electrode active material layer 21 may be the same type or different types.

The negative electrode current collector 22 may include, for example, Cu foil, Ni foil, etc. The negative electrode current collector 22 may have a thickness of, for example, 5 to 50 μm.

<Separator layer>
The separator layer 3 is interposed between the positive electrode 1 and the negative electrode 2. The separator layer 3 separates the positive electrode 1 from the negative electrode 2. The separator layer 3 includes a sulfide solid electrolyte. The separator layer 3 may further include a binder. The sulfide solid electrolyte may be the same or different between the separator layer 3 and the positive electrode active material layer 11. The sulfide solid electrolyte may be the same or different between the separator layer 3 and the negative electrode active material layer 21.

<Mixing of negative electrode slurry>
The inspection method in this embodiment predicts whether or not self-discharge (short circuit) will occur in the cell 100 of the all-solid-state battery based on manufacturing parameters in the kneading process of the negative electrode slurry. In order to facilitate understanding of the characteristics of the inspection method in this embodiment, an outline of the kneading process of the negative electrode slurry in a comparative example will be described. Note that kneading means at least one of dispersing and stirring.

Figure 3 is a flow chart showing the procedure for preparing the negative electrode slurry in the comparative example. Hereinafter, steps are abbreviated as "S". First, the negative electrode material for preparing the negative electrode slurry is weighed (S901). Next, the negative electrode slurry is subjected to a first stirring step (S902), a first dispersion step (S903), a second dispersion step (S904), and a second stirring step (S905) in this order. This completes the kneading step of the negative electrode slurry. Note that a portion of the negative electrode material may be different or the same between the first dispersion step and the second dispersion step.

In a series of processes including two dispersion and stirring steps as described above, the manufacturing parameters of the negative electrode slurry that may cause self-discharge of the cell 100 include the viscosity, particle size, temperature, and solid content during the stirring/dispersion steps.

<Machine learning>
The inventors used a machine learning technique to extract, from among the numerous manufacturing parameters described above, which manufacturing parameters contribute to the self-discharge of the cell 100, in other words, which manufacturing parameters have a high correlation with the self-discharge amount of the cell 100. The results of the machine learning may be stored in the storage 513 as a trained model 64.

As estimation models for machine learning, linear regression, logistic regression, support vector machine (SVM), decision tree, random forest, deep learning (neural network), naive Bayes, k-means, principal component analysis (PCA), LightGBM, XGBoost, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial networks (GAN), etc. can be used. Among them, decision tree methods such as random forest, LightBGM, and XGBoost are preferable. Furthermore, gradient boosting decision tree methods (LightBGM, XGBoost, etc.) that can handle missing values are preferable as estimation models used in this embodiment because they can learn even when some types of values of explanatory variables are missing in a pair of explanatory variables (manufacturing parameters of negative electrode slurry) and objective variables (self-discharge amount) used for learning. In this embodiment (the example of FIG. 4 described later), LightBGM was used.

As a method for calculating the contribution (importance) of each explanatory variable from the estimated model calculated by the above method, Feature Importance, Permutation Importance, SHapley Additive exPlanations (SHAP), etc., which can quantify the correlation between explanatory variables and the objective variable, can be used. In this embodiment, Feature Importance is used.

Figure 4 is a diagram summarizing the contribution of the manufacturing parameters of the negative electrode slurry to the self-discharge amount of the cell 100, as determined by machine learning. The top five manufacturing parameters are shown in Figure 4. As shown in Figure 4, the results showed that the solid content rate in the first stirring process had the highest contribution. The second most important manufacturing parameter was the viscosity in the second dispersion process. The third most important manufacturing parameter was the particle size in the second stirring process. The fourth most important manufacturing parameter was the temperature in the second dispersion process. The fifth most important manufacturing parameter was the particle size after the kneading process of the negative electrode slurry was completed (after S905 was performed).

<Actual measurement results>
The results of an actual evaluation of the relationship between the first to third manufacturing parameters of the negative electrode slurry and the self-discharge amount of the cell 100 will be described.

≪Positive electrode≫
The following materials were used for the positive electrode slurry and the positive electrode current collector.

Positive electrode active material: NCA-based positive electrode active material Solid electrolyte: Sulfide solid electrolyte Conductive material: VGCF
Binder: PVDF
Dispersion medium: Butyl butyrate Positive electrode current collector: Al foil Positive electrode slurry was prepared by kneading the positive electrode active material, solid electrolyte, conductive material, binder, and dispersion medium. An ultrasonic homogenizer was used in the dispersion process of the positive electrode slurry. A stirring blade was used in the stirring process of the positive electrode slurry. A coating film was formed by applying the positive electrode slurry to the surface of the positive electrode current collector 12 by the blade method. The coating film was dried at 100°C for 30 minutes on a hot plate. In this way, a positive electrode raw material was manufactured. A positive electrode 1 was cut out from the positive electrode raw material.

≪Negative electrode≫
The following materials were used for the negative electrode slurry and the negative electrode current collector.

Negative electrode active material: LTO particles Solid electrolyte: Li ₂ SP ₂ S ₅ -based sulfide solid electrolyte Conductive material: VGCF
Binder: PVDF
Dispersion medium: Butyl butyrate Negative electrode current collector: Ni foil Negative electrode active material, solid electrolyte, conductive material, binder, and dispersion medium were mixed together to prepare a negative electrode slurry. An ultrasonic homogenizer was used for dispersing the negative electrode slurry. A stirring blade was used for stirring the negative electrode slurry. A coating film was formed by applying the negative electrode slurry to the surface of the negative electrode current collector 22 by the blade method. The coating film was dried at 100°C for 30 minutes on a hot plate. In this way, a negative electrode blank was manufactured. A negative electrode 2 was cut out from the negative electrode blank.

<Separator layer>
The following materials were used for the separator slurry and substrate:

Solid electrolyte: Sulfide solid electrolyte Binder: PVdF
Dispersion medium: Butyl butyrate Substrate: Al foil A separator slurry was prepared by kneading the solid electrolyte, binder, and dispersion medium. An ultrasonic homogenizer was used for the dispersion process of the separator slurry. A stirring blade was used for the stirring process of the separator slurry. A coating film was formed by applying the separator slurry to the substrate surface by the blade method. The coating film was dried at 100°C for 30 minutes on a hot plate. In this way, a separator raw sheet was manufactured. A separator layer 3 was cut out from the separator raw sheet.

<Battery Construction>
The negative electrode 2 was placed on one side of the separator layer 3 and pressed at 1 ton/cm ² , and then the substrate of the separator layer 3 was removed. Next, the positive electrode 1 was placed on the other side of the separator layer 3 and pressed at 3 ton/cm ² . The electricity storage element and the tab with the welding tape were ultrasonically bonded. The electricity storage element was then sealed with an aluminum laminate film to produce a small laminate cell for an all-solid-state battery.

≪Judgment criteria≫
The voltage of the cell 100 of the all-solid-state battery that had been initially charged was measured after it had been left to stand for two days. The voltage of the cell 100 was measured after it had been left to stand for another day. The difference between these two voltages was taken as the self-discharge amount. If the self-discharge amount of the cell 100 is 3.4 mV or less, the cell 100 is determined to be a good product. On the other hand, if the self-discharge amount of the cell 100 is more than 3.4 mV, the cell 100 is determined to be a defective product.

<Comparison of solid content, viscosity, and particle size>
The evaluation results of the solid content, viscosity, and particle size of the negative electrode slurry with respect to the self-discharge amount will be described. The solid content, viscosity, and particle size can be measured using a moisture meter, an E-type viscometer, and a particle size gauge, respectively. The viscosity was measured when the shear rate was 2 [s ⁻¹ ], 19.2 [s ⁻¹ ], 38.3 [s ⁻¹ ], and 384 [s ⁻¹ ].

FIG. 5 is a diagram showing the relationship between the solid content rate (first explanatory variable) and the self-discharge amount (objective variable) in the first stirring step of the negative electrode slurry. The horizontal axis represents the solid content rate in the first stirring step of the negative electrode slurry. FIG. 6 is a diagram showing the relationship between the viscosity (second explanatory variable) and the self-discharge amount in the second dispersion step of the negative electrode slurry. The horizontal axis represents the viscosity in the second dispersion step of the negative electrode slurry when the shear rate is 38.3 [s ⁻¹ ]. FIG. 7 is a diagram showing the relationship between the particle size (third explanatory variable) and the self-discharge amount in the second stirring step of the negative electrode slurry. The horizontal axis represents the particle size in the second stirring step of the negative electrode slurry. The vertical axis represents the self-discharge amount of the cell in both cases.

As shown in FIG. 5, a correlation was confirmed between the solid content rate in the first stirring step of the negative electrode slurry and the self-discharge amount. From FIG. 5, it can be seen that the self-discharge amount is 3.4 mV or less when the solid content rate in the first stirring step is in the range of more than 43.30% and less than 43.65%. In this range of solid content rate, the self-discharge amount can be predicted from the solid content rate, so the quality of the cell 100 can be determined from the solid content rate.

In contrast, as shown in Figures 6 and 7, no clear correlation was found between the viscosity of the negative electrode slurry in the second dispersion process and the self-discharge amount, nor between the particle size of the negative electrode slurry in the second stirring process and the self-discharge amount. Therefore, the self-discharge amount cannot be predicted from the viscosity in the second dispersion process and the particle size in the second stirring process, and the quality of the cell 100 cannot be determined.

<Processing flow>
8 is a flow chart showing a procedure for preparing the negative electrode slurry in the embodiment. At least steps S103 to S105 and S109 of the steps described below are stored in the control program 62.

First, the negative electrode material for preparing the negative electrode slurry is weighed (S101). Then, the first stirring step of the negative electrode slurry is performed (S102). These steps are the same as steps S901 and S902 in the comparative example (see FIG. 3).

In S103, the server 51 (processor 511) acquires the measurement result of the solid content rate in the first stirring process of the negative electrode slurry. The measurement result of the solid content rate is preferably stored in the battery management data 63. For example, a signal indicating the measurement result of the solid content rate may be output from the moisture meter to the server 51, and the measurement result of the solid content rate may be automatically stored in the battery management data 63. Alternatively, the operator may input the measurement result of the solid content rate by the moisture meter using the input device 52, and the input value may be stored in the battery management data 63. In these cases, the server 51 can acquire the measurement result of the solid content rate in the first stirring process by reading out the value stored in the battery management data 63.

In S104, the server 51 determines whether the measurement result of the solid content rate of the negative electrode slurry in the first stirring process obtained in S103 is within a predetermined reference range. The reference range is determined based on the results of a previous experiment. In the above example, the reference range is a range of solid content rate greater than 43.30% and less than 43.65% (see FIG. 5). The server 51 may display the determination result on the display 53. The server 51 may transmit the determination result to an external device (such as a terminal operated by an operator) via the communication device 54.

If the measurement result of the solid content rate in the first stirring process of the negative electrode slurry is within the standard range (YES in S104), the server 51 determines that the negative electrode slurry is suitable for use (to manufacture the cell 100) (S105). In this case, the first dispersion process, the second dispersion process, and the second stirring process are performed on the negative electrode slurry (S106 to S108). These processes are equivalent to the processes S903 to S905 in the comparative example.

On the other hand, if the measurement result of the solid content rate in the first stirring process of the negative electrode slurry is outside the reference range (NO in S104), the server 51 determines that the negative electrode slurry is not suitable for use (S109). In this case, further dispersion and stirring processes are not performed on the negative electrode slurry. The negative electrode slurry is, for example, discarded.

As described above, in this embodiment, based on the knowledge that there is a correlation between the solid content rate in the first stirring process, which is one of the manufacturing parameters of the negative electrode slurry, and the self-discharge amount of the cell 100 (see FIG. 5), it is determined whether the solid content rate in the first stirring process is within the standard range. While the negative electrode slurry whose solid content rate in the first stirring process is within the standard range is passed to the subsequent kneading process, the negative electrode slurry whose solid content rate in the first stirring process is outside the standard range is determined to be unsuitable for incorporation into the cell 100. This makes it possible to determine whether the cell 100 is good or bad before assembling the cell 100. Therefore, according to this embodiment, it is possible to efficiently inspect the cell 100 of the all-solid-state battery.

In this embodiment, it has been described that the kneading process of the negative electrode slurry includes two dispersion steps and two stirring steps. However, the kneading process may include at least one dispersion step and one stirring step. The kneading process may include one dispersion step and one stirring step, one dispersion step and one stirring step, and two other steps, or three or more dispersion steps and stirring steps.

The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is indicated by the claims rather than the description of the embodiments above, and is intended to include all modifications within the meaning and scope of the claims.

100 Cell, 1 Positive electrode, 11 Positive electrode active material layer, 12 Positive electrode current collector, 2 Negative electrode, 21 Negative electrode active material layer, 22 Negative electrode current collector, 3 Separator layer, 5 Inspection system, 51 Server, 511 Processor, 512 Memory, 513 Storage, 514 Network interface, 52 Input device, 53 Display, 54 Communication device, 61 System program, 62 Control program, 63 Battery management data, 64 Trained model.

Claims (3)

A method for inspecting an all-solid-state battery, comprising:
The all-solid-state battery includes a plurality of coating layers,
Each of the plurality of coating layers is formed by coating a current collector or a substrate with a slurry prepared through a kneading process of a solid electrolyte and a solvent,
The plurality of coating layers are
a positive electrode layer formed by coating a positive electrode slurry;
a negative electrode layer formed by coating a negative electrode slurry containing a negative electrode active material, a negative electrode solid electrolyte, a conductive material, a binder, and a dispersion medium ;
a separator layer formed by coating the separator slurry;
The inspection method includes:
acquiring a measurement result of a solid content rate of the negative electrode layer in a first stirring step of the negative electrode slurry;
determining that the negative electrode layer is not suitable for use when the obtained solid content rate is outside a reference range defined based on a correspondence relationship between the solid content rate of the negative electrode slurry and a self-discharge amount of the all-solid-state battery;
The negative electrode active material includes LTO particles,
The negative electrode solid electrolyte includes a Li ₂ S—P ₂ S ₅ -based sulfide solid electrolyte,
The conductive material includes VGCF,
The binder comprises PVdF,
The method for inspecting an all-solid-state battery, wherein the dispersion medium includes butyl acetate . A manufacturing parameter of the slurry that contributes to a self-discharge amount of the all-solid-state battery is extracted using a trained model by supervised learning;
The trained model has a decision tree,
2 . The method for inspecting an all-solid-state battery according to claim 1 , wherein the decision tree includes the self-discharge amount as a response variable and a solid content rate in a first stirring step of the negative electrode slurry as an explanatory variable. one or more processors;
A memory.
The memory causes the one or more processors to execute the inspection method according to claim 1 or 2 by executing a program by the one or more processors.

Images