Materials innovation support methods, materials innovation support systems, and evaluation devices

By utilizing a materials innovation support system and machine learning and automated experimental systems, the slow speed caused by forming and firing in ceramic materials innovation has been solved, enabling rapid evaluation and synthesis, and improving the innovation speed of ceramic materials, especially the innovation efficiency of functional ceramic materials.

CN123125656APending Publication Date: 2026-11-17NGK CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202480090863.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-04-23
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In ceramic material innovation, existing technologies require each sample to be molded and fired, resulting in slow innovation speed and high time and labor costs. This is especially true in the innovation of functional ceramic materials, which involves multiple levels and makes it difficult to efficiently evaluate their physical properties.

Method used

By leveraging a materials innovation support system and utilizing machine learning models and automated experimental systems, the physical properties of ceramic powders can be predicted, enabling rapid evaluation of ceramic powders, reducing molding and firing steps, and automating the synthesis and evaluation of ceramic raw materials through high-throughput experiments and robotic systems, thus achieving rapid innovation.

Benefits of technology

It improves the speed of innovation in ceramic materials, reduces the time and raw materials required for synthesis processes, and enables rapid prediction of the physical properties of sintered ceramic materials, making it particularly suitable for the rapid innovation of functional ceramic materials.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123125656A_ABST
    Figure CN123125656A_ABST
Patent Text Reader

Abstract

The material innovation support method includes an evaluation process, in which multiple ceramic raw materials are reacted to synthesize ceramic powder, and the physical properties of the ceramic powder are evaluated to predict the crystalline phase and physical properties of the ceramic material after sintering.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention largely relates to support for materials innovation. Background Technology

[0002] As a step in materials innovation, there is the evaluation of materials innovated in the form of samples during experiments. In Non-Patent Literature 1, the material used as the sample is a catalyst, and its catalytic activity is evaluated.

[0003] Existing technical documents

[0004] Non-patent literature

[0005] Non-patent document 1: Masanori Kodera and Kazuhiro Sayama, "An automatic robotsystem for machine learning-assisted high-throughput screening of compositeelectrocatalysts", Digital Discovery, 2023, 2, 1683-1687, published by the Royal Society of Chemistry, on 7 October 2023 Summary of the Invention

[0006] In the innovation of ceramic materials, the evaluation of the crystalline phase and physical properties of ceramic materials is carried out.

[0007] In this specification, the term "crystalline phase" refers to the order in which the constituent atoms are arranged and whether they possess symmetry, but does not include the property of how they respond to external factors (typically physical forces, heat, electricity, magnetism, or light). "Crystalline phase" can also be referred to as a "material characteristic," which refers to a physical and / or chemical state. Specifically, it may include, for example, one or more characteristic items (e.g., grain size, grain size distribution, etc.) and values ​​associated with each of those characteristic items.

[0008] In this specification, the term "physical property" refers to the property of how a substance responds to external factors. "Physical property" can also be called "material properties," which refers to the physical and / or chemical properties exhibited by a material. Specifically, it may include one or more property items (such as rigidity, thermal conductivity, electrical conductivity, dielectric constant, or magnetic susceptibility) and values ​​(typically numerical values) associated with each of these property items.

[0009] Furthermore, the term "ceramics" refers to a general term for inorganic materials composed of polycrystalline materials, which may include metals and organic compounds. Additionally, ceramics can be composed of single crystals or amorphous materials such as glass. In this specification, "ceramics" can refer to polycrystalline, single-crystalline, or amorphous materials, and these can also be mixed.

[0010] As a ceramic material, it has two categories: structural ceramic materials, which are obtained by mixing and firing commercially available raw materials and do not require the synthesis of raw materials; and functional ceramic materials, which require the synthesis of raw materials.

[0011] Regarding functional ceramic materials, their properties vary depending on the additives used or their ratios. Therefore, the innovation of functional ceramic materials (such as composite oxides or composite nitrides) involves multiple levels of prototyping. For example, ABO3 is a functional ceramic material. Within ABO3, compounds doped with various elements at various ratios at different sites need to be synthesized, and prototyping involves multiple levels. To evaluate the properties of ceramic materials for each sample (each ceramic powder), a large number of samples need to be molded and fired separately. Furthermore, to implement molding and firing, a large number of samples need to be synthesized. Therefore, the innovation of functional ceramic materials is time-consuming and labor-intensive.

[0012] In Non-Patent Literature 1, the material being innovated is a catalyst, and the evaluation focuses on the material itself. Therefore, even if the technology disclosed in Non-Patent Literature 1 is applied to the innovation of ceramic materials, it is still necessary to form and fire each sample, and the aforementioned problems remain unresolved.

[0013] The material innovation support method includes an evaluation process, in which multiple ceramic raw materials are reacted to synthesize ceramic powder, and the physical properties of the ceramic powder are evaluated to predict the crystalline phase and physical properties of the ceramic material after sintering.

[0014] Invention Effects

[0015] It can accelerate the innovation of ceramic materials. Attached Figure Description

[0016] Figure 1 An example of the system configuration in one embodiment of the present invention is shown.

[0017] Figure 2 This illustrates the composition of the experimental system and an example of the processing flow performed by the materials innovation support system.

[0018] Figure 3 Show Figure 2 A specific example. Detailed Implementation

[0019] In the following description, "interface device" can refer to more than one interface device. The more than one interface device can be at least one of the following.

[0020] • One or more I / O (Input / Output) interface devices. An I / O (Input / Output) interface device is an interface device for at least one of an I / O device and a computer for remote display. The I / O interface device for the display computer can be a communication interface device. At least one I / O device can be any of a user interface device, such as an input device like a keyboard and click device, and an output device like a display device.

[0021] • More than one communication interface device. More than one communication interface device can be more than one of the same type of communication interface device (e.g., more than one NIC (Network Interface Card)) or more than two different types of communication interface devices (e.g., NIC and HBA (Host Bus Adapter)).

[0022] Additionally, in the following description, "memory" refers to one or more memory devices as an example of more than one memory device, typically a main memory device. At least one memory device in the memory can be a volatile memory device or a non-volatile memory device.

[0023] Additionally, in the following description, "permanent storage device" can refer to one or more permanent storage devices as an example of more than one storage device. Permanent storage devices are typically non-volatile storage devices (e.g., secondary storage devices), specifically, such as HDD (Hard Disk Drive), SSD (Solid State Drive), NVME (Non-Volatile Memory Express) drives, or SCM (Storage Class Memory).

[0024] Additionally, in the following description, "storage device" can refer to at least memory in both memory and permanent storage devices.

[0025] Additionally, in the following description, "processor" can refer to more than one processor device. At least one processor device can typically be a microprocessor device such as a CPU (Central Processing Unit), but can also be other types of processor devices such as a GPU (Graphics Processing Unit). At least one processor device can be single-core or multi-core. At least one processor device can be a processor core. At least one processor device can be a processor device in a broad sense, which is a collection of gate arrays (e.g., FPGA (Field-Programmable Gate Array), CPLD (Complex Programmable Logic Device), or ASIC (Application Specific Integrated Circuit)) constructed using hardware description languages ​​that perform some or all of the processing.

[0026] Furthermore, in the following description, functions are sometimes described using the expression "yyy part." A function can be implemented by executing one or more computer programs using a processor, by implementing one or more hardware circuits (e.g., FPGA or ASIC), or by a combination thereof. When a function is implemented by executing a program using a processor, the prescribed processing is performed while appropriately employing storage devices and / or interface devices, therefore, the function can be at least a part of the processor. The processing described with the function as the subject can be processing performed by the processor or a device having the processor. The program can be installed from a program source. The program source can be, for example, a program distribution computer or a recording medium that a computer can read (e.g., a non-temporary recording medium). Each function is described as an example; multiple functions can be combined into one function, or one function can be divided into multiple functions.

[0027] Additionally, in the following descriptions, when no similar elements are distinguished in the description, the general reference mark is used; when similar elements are distinguished, the reference mark is sometimes used.

[0028] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. It should be noted that in the following description, "MI" is an abbreviation for Materials Informatics and "DB" is an abbreviation for Database.

[0029] Figure 1An example of the system configuration in one embodiment of the present invention is shown. It should be noted that at least a portion of the descriptions disclosed in the prior applications PCT / JP2023 / 009174 and PCT / JP2023 / 009175 of the same applicant as this application can be referenced in this description. Hereinafter, a brief description will be provided. Figure 1 Explanation.

[0030] The materials innovation support system 10 includes: an experimental system 110, a data acquisition unit 60, an MI platform system 100, and a researcher terminal 11A. The MI platform system 100 includes a data prompting unit 50 and a data conversion unit 70.

[0031] The data prompting unit 50 has an AI (Artificial Intelligence) unit 108 and an IF (Interface) unit 109. The AI ​​unit 108 learns a machine learning model and uses the inferences derived from the model. Responding to instructions, for example, from the IF unit 109, the AI ​​unit 108 inputs processed data obtained from the data mart 107 into the machine learning model, thereby obtaining an inference result, and outputs the inference result to the IF unit 109. The IF unit 109 receives inquiries from the researcher terminal 11A and, in response, prompts the researcher terminal 11A with object data. The researcher terminal 11A is an information processing terminal (e.g., a personal computer or smartphone) of the materials researcher 5A (an example of a user). The researcher terminal 11A is an example of a sender of object data inquiries, and also an example of a prompter of object data. The IF unit 109 prompts the object represented by the object data (processed data obtained from the data mart 107, and / or data based on the inference results obtained from instructions in the AI ​​unit 108) when it receives inquiries from the researcher terminal 11A (or periodically). The object being suggested can be, for example, a manufacturing process or material properties. In this embodiment, "manufacturing process" refers to an innovative method for producing materials, which typically includes material composition and / or synthesis process. "Material composition" can be, for example, a blending composition, and "synthesis process" can be the type of raw materials (e.g., particle size differences), process conditions, etc. The manufacturing process may also include a material development strategy.

[0032] Materials researcher 5A uses the provided object data as a basis to innovate new materials or conduct experiments based on the object data. The party providing the object data (sender) can be a system such as experimental system 110, instead of researcher terminal 11A or other than researcher terminal 11A.

[0033] Based on the proposed object, an experiment using experimental system 110 is conducted. Experimental system 110 may include: fabrication system 111 (e.g., an atmospheric firing furnace), robotic system 187 for fabrication system 111 and / or evaluation system 112, and evaluation system 112 (e.g., a device for evaluating the rate of thermal expansion). Experimental data is output from these experimental devices, fabrication system 111 and evaluation system 112. The output experimental data is sent to data server 120 and stored. Alternatively, a combined experiment can be used, or in addition to a combined experiment, at least one of the following can be employed: high-throughput experiment, automated experiment using robots, and experiment primarily based on manual operation. Fabrication system 111 and robotic system 187 can be operated using MI platform system 100 according to the manufacturing formula data.

[0034] The data acquisition unit 60 includes a data server 120 and an experiment note unit 130. The data server 120, as a database storing experimental data from the experiment system 110, includes a production database 121 storing experimental data from the production system 111, and an evaluation database 122 storing experimental data from the evaluation system 112. "Experimental data" can be data related to the experiment, such as a summary, details, results, or evaluations. The experiment note unit 130 manages the experimental data obtained from the databases managed by the data server 120 as electronic experiment notes 131.

[0035] The data conversion unit 70 includes a collection unit 101, a feature calculation unit 103, an image analysis unit 104, a natural language analysis unit 105, and a processing unit 106. The collection unit 101 collects experimental data from the electronic lab notebook 131 and the database of the data server 120, shapes the collected experimental data, and stores the shaped experimental data in the data lake 102. Alternatively, it collects various types of data 150 and stores at least a portion of these types of data in the data lake 102. The feature calculation unit 103 calculates the feature quantities of one or more types of data specified in the shaped experimental data. For example, the one or more types of data can be image data and text data. Image data is analyzed by the image analysis unit 104, and based on the analysis result, the feature calculation unit 103 calculates the feature quantities of the image data. Additionally, text data is text-mined by the natural language analysis unit 105, and based on the text-mining result, the feature calculation unit 103 calculates the feature quantities of the text data. The sorting department 106 sorts the data lake 102 into one or more data marts 107 that are data sets that meet the specified conditions.

[0036] according to Figure 1The system shown can be expected to efficiently and quickly perform the following cycle: object data prompting → experiment → experimental data collection → conversion of experimental data into organized data → object data prompting based on the organized data. For example, based on the organized data collected and shaped by the MI platform system 100, the object data is prompted to materials researcher 5A using the MI platform system 100. Alternatively, data can be prompted to materials researcher 5A without going through the MI platform system 100. For example, another materials researcher 5B can obtain experimental data from electronic lab notebook 131 using trainer terminal 11B, organize the experimental data, and input it into AI unit 140. AI unit 140 can also be implemented outside (or within) trainer terminal 11B.

[0037] The MI platform system 100 provides ceramic manufacturing formula data as object data, representing the ceramic manufacturing formula. According to the manufacturing formula represented by the manufacturing formula data, ceramics (ceramic materials) are generated and experiments are conducted.

[0038] The production of ceramics according to the formulation may include the synthesis (blending) of ceramic raw materials. At least a portion of the synthesis of ceramic raw materials can be performed manually (e.g., by materials researcher 5A). However, in this embodiment, the MI platform system 100 (an example of a computer system) can operate the manufacturing system 111 and the robot system 187 according to the formulation data, thereby automatically performing all or part of the ceramic raw material synthesis process. This allows for high-speed implementation of the ceramic raw material synthesis process. It should be noted that the synthesis process may include multiple steps, and at least one of these steps can be performed automatically. Furthermore, the MI platform system 100 can receive the formulation data from an external system instead of generating it, and operate the manufacturing system 111 and the robot system 187 according to the received formulation data. That is, a computer system like the MI platform system 100 can generate or receive formulation data, and operate the manufacturing system 111 and the robot system 187 according to the generated or received formulation data. The synthesis process of ceramic raw materials can, for example, utilize at least a portion of the descriptions disclosed in earlier applications PCT / JP2023 / 031646 and PCT / JP2024 / 012942 by the same applicant as this application.

[0039] Ceramic powder (ceramic raw material powder) is obtained by reacting and synthesizing multiple ceramic raw materials through manufacturing system 111 (or manually). Evaluation system 112 performs an evaluation process. In the evaluation process, evaluation system 112 evaluates the physical properties of the ceramic powder to predict the crystalline phase and physical properties of the sintered ceramic material. In other words, evaluation system 112 evaluates the physical properties of the ceramic powder in order to predict the physical properties of the ceramic material after sintering. Accordingly, the physical properties of the sintered ceramic material can be predicted without the need for shaping and firing the ceramic powder, thus increasing the innovation speed of ceramic materials. In addition, since the shaping and firing of the ceramic powder is not required, the raw materials for synthesis can be reduced. As a result, the time required for the synthesis process can be expected to be reduced, which also suggests an increase in the innovation speed of ceramic materials. For example, in the synthesis process, liquid-phase synthesis can be performed as the synthesis of multiple ceramic raw materials. The amount of liquid in liquid-phase synthesis is less than in the case where the shaping and firing of ceramic powder is required.

[0040] In this embodiment, the physical properties of the ceramic material, which will ultimately be sintered, are predicted at the particle stage. Typical physical properties include mechanical properties, thermal properties, electrical properties, magnetic properties, and optical properties. The thermal, electrical, and magnetic properties of the particles are substantially unaffected by variations in molding or firing conditions. With this in mind, the physical properties of the ceramic powder evaluated in the evaluation process of this embodiment are at least one of thermal, electrical, and magnetic properties. It should be noted that the physical properties evaluated are not limited; for example, mechanical and / or optical properties may be evaluated in addition to at least one of thermal, electrical, and magnetic properties.

[0041] Furthermore, the evaluation system 112 automatically evaluates the physical properties of the ceramic powder when it is provided. Based on this, it is expected that the pace of innovation in ceramic materials will be further accelerated.

[0042] Furthermore, the ceramic material in this embodiment can be a functional ceramic material. Functional ceramic materials differ from structural ceramic materials in that they are ceramic materials that require raw material synthesis (the prototype involves multiple layers). If the ceramic material in this embodiment is a functional ceramic material, the innovation speed of such multi-layered ceramic materials can be increased.

[0043] The following is a detailed description of this embodiment.

[0044] Figure 2 This illustrates the configuration of the experimental system 110 and an example of the processing flow performed by the material innovation support system. Figure 3 Show Figure 2 A specific example.

[0045] In addition to the evaluation process, the material innovation support method may also include a synthesis process and / or a composition determination process. In this embodiment, if the physical properties evaluated in the evaluation process do not meet the target physical properties, a composition determination process is performed. Then, in the synthesis process, the material is synthesized according to the determined composition, and the physical properties of the resulting ceramic powder are evaluated again. That is, the cycle of evaluation process → composition determination process → synthesis process → evaluation process is repeated until an evaluation meeting the target physical properties is obtained.

[0046] Specifically, the material innovation support system 10 performs the following steps (a) to (c) (using the composition determined in (c) as the specified composition, (a) is performed again). Accordingly, the composition of ceramic powder 290 with properties matching the target properties can be quickly found, thereby contributing to the improvement of the innovation speed of ceramic materials.

[0047] (a) Performing a synthesis process, which includes: synthesizing multiple ceramic raw materials 211 according to a specified composition.

[0048] (b) An evaluation process is performed, which includes evaluating the physical properties of the ceramic powder 290 obtained by the synthesis process in (a).

[0049] (c) Perform a composition determination process, in which different compositions are determined based on accumulated data, including data as evaluation results in (b), if the predicted physical properties do not meet the target physical properties.

[0050] The data representing the evaluation results in (b) is accumulated in the evaluation DB122. In (c), the MI platform system 100 determines whether the predicted physical properties represented by the evaluation result data obtained from the evaluation DB122 conform to the target physical properties. If the determination result is false, the AI ​​unit 108 of the MI platform system 100 uses the data accumulated in the evaluation DB122 to update the machine learning model 210, which outputs data representing composition based on the input data representing physical properties. The updated machine learning model 210 is then input with data representing the target physical properties, thereby outputting data representing the composition (at least a part of the manufacturing formula) of the ceramic material expected to have the target physical properties. The manufacturing formula data based on this data is output using the IF unit 109, and the synthesis process is performed using the manufacturing system 111 according to the composition represented by the output manufacturing formula data. Accordingly, the innovation speed of ceramic materials can be improved.

[0051] The machine learning model 210 can be any model, such as linear regression, logistic regression, decision tree model, neural network, Bayesian optimization, k-NN, or a combination of any two or more of these models. In this embodiment, the machine learning model 210 is a model with physical properties as the target variable and multiple composition-related items as explanatory variables. Composition-related items can be elements or their ratios. For example, in the case of a ceramic material of ABO3, composition-related items can be elements doped at site A or their ratios, elements doped at site B or their ratios, etc. Updating the machine learning model 210 based on the data accumulated in the evaluation DB122 can include updating the coefficients of the explanatory variables.

[0052] In (b), the evaluation system 112 can evaluate not only the physical properties of the ceramic powder obtained through the synthesis process, but also the composition of the ceramic powder. For example, the evaluation system 112 may include, as an example, an evaluation device 202 for evaluating composition, such as ICP-MS (Inductively Coupled Plasma Mass Spectrometry) and / or a fluorescence X-ray analysis device. Using such an evaluation device 202, the composition of the ceramic powder can be evaluated by identifying and quantifying the types of constituent ions. The composition evaluation may include confirmation that the ceramic raw materials were synthesized in the synthesis process according to the specified composition. The data representing the evaluation results in (b) may include data representing one or more sets of evaluated composition and predicted physical properties. The machine learning model 210 can be updated based on data representing one or more sets of evaluated composition and predicted physical properties. Accordingly, the machine learning model 210 is updated based on a more accurate relationship between composition and physical properties, thus increasing the possibility of rapidly obtaining the composition output of ceramic materials with target physical properties.

[0053] In this embodiment, the cycle from (a) to (c) described above is performed automatically as follows. The cycle ends when the predicted physical properties match the target physical properties. Accordingly, the innovation speed of ceramic materials can be improved.

[0054] In (a), the MI platform system 100 controls the production system 111 for the synthetic ceramic raw materials, thereby automatically performing the synthesis according to the composition of the preparation formula.

[0055] In (b), by utilizing the MI platform system 100 to operate the robot system 187 and the evaluation system 112, the following operations are automatically performed: the ceramic powder synthesized in (a) is automatically placed into the evaluation system 112, the physical properties of the placed ceramic powder are automatically evaluated, and the ceramic powder whose physical properties have been evaluated is automatically removed from the evaluation system 112. The robot system 187 includes one or more robots, which are used to place samples into the evaluation system 112 and remove samples from the evaluation system 112.

[0056] In (c), the MI platform system 100 automatically performs the following: determining whether the predicted physical properties conform to the target physical properties, and determining different compositions when the predicted physical properties do not conform to the target physical properties. Based on the compositions determined in (c), (a) is performed automatically again.

[0057] The manufacturing system 111 performs a synthesis process. In this process, the manufacturing system 111 synthesizes multiple ceramic raw materials (e.g., raw materials 211A to 211C) according to a manufacturing formula (an example of composition) expressed in the manufacturing formula data output from the MI platform system 100 (an example of a computer system), thereby obtaining ceramic powder 290. The manufacturing system 111 may include multiple devices, which can be used to perform multiple steps constituting the synthesis process (e.g., weighing, raw material mixing, drying, crushing, and separation). Manufacturing formula data can be input into the manufacturing system 111 from the IF unit 109 of the MI platform system 100, and the synthesis process according to the composition expressed in the manufacturing formula data is performed under the control of the MI platform system 100 and the robot system 187.

[0058] Robot system 187 may include one or more robots. The robots in robot system 187 may be collaborative robots or other robots (e.g., industrial robots). Furthermore, the robots may be prepared separately for at least one of one or more devices in manufacturing system 111, and / or at least one of one or more evaluation devices 202 in evaluation system 112, or a single robot may be used in two or more devices in manufacturing system 111, and / or two or more evaluation devices 202 in evaluation system 112. The robots in robot system 187 receive requests directly or indirectly from MI platform system 100 via, for example, a communication network, and perform actions according to the received requests. In the case of indirect request acceptance, a control device for at least one robot may receive the request from MI platform system 100 and send a request based on that request to the at least one robot. In the case of direct request acceptance, the request may be received from MI platform system 100 without such a control device. Hereinafter, for the sake of simplicity, regarding any robot in robot system 187, whether the request is received directly or indirectly, it receives a request from MI platform system 100 according to the manufacturing formula data 200. Furthermore, even if any one of the robots in the robot system 187 performs an action, the robot system 187 will still be activated. The robot can send a notification to the MI platform system 100 at the start or completion of the requested action (e.g., a notification of status such as completion of the request).

[0059] like Figure 2 As shown, the evaluation system 112 includes one or more evaluation devices 202. Using a robotic system 187 (or manually), a sample of ceramic powder 290 is placed in one or more evaluation devices 202, and each of the one or more evaluation devices 202 evaluates the sample. The evaluation database 122 accumulates evaluation result data, including data representing the evaluation results of each evaluation device 202. The accumulation of evaluation result data in the evaluation database 122 can be performed automatically by the evaluation system 112, with or without the MI platform system 100. The evaluation result data can be data for each sample, which may include data representing the physical properties of the sample (e.g., physical property values ​​obtained for multiple physical property items such as coefficient of thermal expansion and electrical conductivity). In addition to data representing the physical properties of the sample, the evaluation result data for each sample may also include data representing the evaluated composition of the sample.

[0060] The MI platform system 100 determines whether the predicted physical properties represented by the accumulated data in DB122 conform to the target physical properties. When the target physical properties and predicted physical properties include multiple items (such as coefficient of thermal expansion and electrical conductivity), the determination that the predicted physical properties conform to the target physical properties can be: for all of the multiple items, the values ​​in the predicted physical properties conform to the target values ​​in the target physical properties.

[0061] In the evaluation system 112, as an example of the evaluation device 202, such as Figure 3 As shown, it may include: a high-temperature XRD device 202A and an sMIM (scanning microwave impedance microscope) 202B.

[0062] The high-temperature XRD apparatus 202A evaluates the coefficient of thermal expansion (an example of a thermal property) using high-temperature XRD. The high-temperature XRD apparatus 202A can perform high-temperature XRD measurements automatically under control (or without control) from the MI platform system 100 and / or the robot system 187. The evaluation process may include, for example, performing high-temperature XRD measurements using the high-temperature XRD apparatus 202A automatically: placing a sample into the high-temperature XRD apparatus 202A, heating the placed sample to a predetermined temperature, performing XRD measurements on the sample at the predetermined temperature, and removing the sample from the high-temperature XRD apparatus 202A. Specifically, for example, the high-temperature XRD apparatus 202A and the robot in the robot system 187 can cooperate to automatically perform: placing a sample into the high-temperature XRD apparatus 202A, heating the placed sample to a predetermined temperature, performing XRD measurements on the sample at the predetermined temperature, and removing the sample from the high-temperature XRD apparatus 202A. For example, an automatic sample changer may be integrated into the high-temperature XRD apparatus 202A. The automatic sample converter can be an element included in the robot system 187, or it can be a separate element from the robot system 187. Multiple samples (ceramic powders) of different compositions can be mounted on the automatic sample converter using, for example, the robot system 187, to evaluate the coefficients of thermal expansion of the multiple samples, and the data representing the coefficients of thermal expansion obtained for each of the multiple samples can be accumulated in the evaluation DB122 as at least a part of the evaluation result data.

[0063] The sMIM202B is used to evaluate conductivity. Although not illustrated, conductivity can be evaluated using impedance measurement or spectrophotometry, instead of the sMIM202B. For each sample, the conductivity data obtained from the evaluation, along with, for example, the thermal expansion coefficient data of the sample, can be accumulated in the evaluation DB122 as at least a portion of the evaluation result data.

[0064] At least one evaluation device 202, such as a high-temperature XRD device 202A, may include: an input section for automatically feeding ceramic powder; a temperature control section for automatically heating the fed ceramic powder to a predetermined temperature; a measurement section for automatically performing XRD measurements on the ceramic material at the predetermined temperature; and a removal section for automatically removing the ceramic powder from the high-temperature XRD device 202A. The input and removal sections can be implemented using a robot such as an automated sample changer. The combination of such an evaluation device 202 and the MI platform system 100 enables an increase in the speed of innovation in ceramic materials.

[0065] The above description describes one embodiment; however, this is merely an example to illustrate the invention and does not limit the scope of the invention to this embodiment. The invention can also be implemented in various other forms.

[0066] Explanation of reference numerals in the attached figures

[0067] 10…Materials Innovation Support System, 100…MI Platform System, 112…Evaluation System, 187…Robotic System

Claims

1. A method for supporting material innovation, characterized in that, The physical properties of ceramic powder obtained by the reaction synthesis of multiple ceramic raw materials are evaluated, thereby predicting the physical properties of ceramic materials sintered using the ceramic powder.

2. The material innovation support method according to claim 1, characterized in that, The evaluation of the ceramic powder includes an automatic assessment of its physical properties.

3. The material innovation support method according to claim 1, characterized in that, (a) A process for synthesizing ceramic powder, comprising: reacting and synthesizing multiple ceramic raw materials according to a specified composition. (b) An evaluation process, which includes evaluating the physical properties of the ceramic powder obtained using the synthesis process in (a). (c) In cases where the predicted physical properties do not conform to the target physical properties, different compositions are determined based on accumulated data, including the data used as evaluation results in (b). The components determined in (c) are used as the specified components, and (a) is performed.

4. The material innovation support method according to claim 3, characterized in that, In (c), the accumulated data is used to update the machine learning model that outputs data representing composition in response to input data representing physical properties. The updated machine learning model is then input with data representing the target physical property, thereby outputting data representing the composition of a ceramic material expected to possess that target physical property. The data used as the evaluation result in (b) includes data representing the predicted physical properties. The output data represents the different compositions.

5. The material innovation support method according to claim 4, characterized in that, In (b), in addition to utilizing the physical properties of the ceramic powder obtained from the synthesis process, the composition of the ceramic powder is also evaluated. The data that serves as the evaluation result in (b) includes data representing one or more groups of the evaluated components and the predicted physical properties.

6. The material innovation support method according to claim 1, characterized in that, The physical property is at least one of thermal properties, electrical properties, and magnetic properties.

7. The material innovation support method according to claim 6, characterized in that, In the evaluation process, the coefficient of thermal expansion is evaluated using high-temperature XRD.

8. The material innovation support method according to claim 6, characterized in that, In the evaluation process, conductivity is evaluated using sMIM, impedance measurement, or spectrophotometry.

9. The material innovation support method according to claim 1, characterized in that, The synthesis of the multiple ceramic raw materials is a liquid-phase synthesis.

10. The material innovation support method according to claim 5, characterized in that, The evaluation process includes high-temperature XRD measurement using a high-temperature XRD device, which is automated and involves: placing a sample into the high-temperature XRD device; heating the placed sample to a specified temperature; performing XRD measurement on the sample that has reached the specified temperature; and removing the sample from the high-temperature XRD device.

11. The material innovation support method according to claim 3, characterized in that, The following will automatically loop from (a) to (c). In (a), a computer system is used to control the production system of synthetic ceramic raw materials, thereby automatically performing the synthesis according to the specified composition. In (b), a computer system is used to operate a robotic system, including one or more robots that perform the functions of placing samples into and removing samples from the evaluation system, as well as the evaluation system itself. This allows the ceramic powder synthesized in (a) to be automatically placed into the evaluation system, its properties to be automatically evaluated, and the evaluated ceramic powder to be automatically removed from the evaluation system. In (c), the computer system automatically performs the following: determining whether the predicted physical property conforms to the target physical property, and determining the different compositions when the predicted physical property does not conform to the target physical property. The components determined in (c) are used as the specified components, and (a) is performed automatically. The loop ends when the predicted physical properties match the target physical properties.

12. The material innovation support method according to claim 1, characterized in that, The ceramic material is a functional ceramic material.

13. An evaluation device, characterized in that, have: The feeding section automatically feeds in ceramic powder obtained from the synthesis of multiple ceramic raw materials; The evaluation department automatically evaluates the physical properties of the ceramic powder to predict the physical properties of the ceramic material after sintering; and The extraction section automatically extracts the evaluated ceramic powder.

14. The evaluation device according to claim 13, characterized in that, The evaluation device is a high-temperature XRD device.

15. A material innovation support system, characterized in that, have: An evaluation system is used to evaluate the physical properties of ceramic powder after sintering ceramic powder synthesized by reacting multiple ceramic raw materials according to a specified composition, predicting the crystalline phase and physical properties of the ceramic material. as well as A computer system determines whether a predicted physical property conforms to a target physical property. If the predicted physical property does not conform to the target physical property, the system updates a machine learning model that outputs data representing composition data in response to input data representing the physical property, using accumulated data including data as evaluation results obtained from the evaluation system. The updated machine learning model is then input with data representing the target physical property, thereby outputting data representing the composition of a ceramic material expected to possess the target physical property. The data that serves as a result of the evaluation includes data representing the predicted physical properties. The output data represents the specified composition.

16. The material innovation support system according to claim 15, characterized in that, It also has: A manufacturing system that synthesizes multiple ceramic raw materials according to the composition represented by data output from the computer system. The evaluation system evaluates the physical properties of the ceramic powder, which is the result of the synthesis by the manufacturing system.