Contact angle prediction method, contact angle prediction device, contact angle prediction model generation method, and contact angle prediction model generation device

Abstract

This makes it possible to easily predict the contact angle between a solid surface and molten metal. [Solution] For each combination of solid and liquid, the processor 11 acquires and analyzes information on the contact angle of the liquid on the solid surface, as well as the melting point and thermodynamic parameters related to the solid and liquid, thereby acquiring a prediction model 122 that outputs a predicted value of the contact angle. The melting point information is the melting point information of the solid or liquid, and the thermodynamic parameter information includes the thermodynamic parameters of stable compounds among the compounds composed of the elements that make up the solid or liquid.

Description

Technical Field

[0001] The present invention particularly relates to a contact angle prediction method, a contact angle prediction apparatus, a contact angle prediction model generation method, and a contact angle prediction model generation apparatus regarding the contact angle between a solid surface and a molten metal.

Background Art

[0002] The wettability of a liquid with respect to a solid is an important factor in, for example, the steelmaking process. In the steelmaking process, phenomena that can be explained by the wettability of a liquid with respect to a solid surface, specifically a molten metal, include the separation phenomenon of deoxidation products in the steelmaking deoxidation reaction, for example, oxides derived from deoxidation elements and molten steel, the erosion phenomenon of refractories by molten steel, and non-plating, which is a defect that appears when applying molten plating to a steel plate.

[0003] Since the above phenomena are basically phenomena at high temperatures, it has been necessary to spend a great deal of time and effort to directly measure the contact angle, which is one of the evaluation indices of wettability. Therefore, attempts have been made to predict the contact angle by calculation.

[0004] For example, Non-Patent Document 1 describes that the greater the difference between the free energy change of formation of an oxide in the solid state and the free energy change of the liquid metal (molten metal) to form the oxide, the smaller the contact angle. Also, Non-Patent Document 2 describes that the contact angle can be organized from thermodynamic quantities obtained from a solid, a molten metal, and a combination of a solid / molten metal, such as an equilibrium constant and an activity. Further, Non-Patent Document 3 proposes a contact angle prediction model in a molten metal / oxide system with the physical properties of a metal (atomic weight, density, melting point Tm, surface energy under experimental temperature and oxygen partial pressure, and equilibrium partial pressure PM), the physical properties of an oxide (molecular weight, surface energy, free energy of formation, band gap energy, dissolution energy / RT, and equilibrium oxygen partial pressure PO), experimental conditions (temperature T, oxygen partial pressure P), and mixing characteristics (the ratio of T / Tm and P / PM) as explanatory variables and the reported contact angle as the objective variable.

Prior Art Documents

[0005] [Non-Patent Document 1] YV Naidich: Progress in Suface and Membrane Science, 14(1981) 353. [Non-Patent Document 2] N. Eustathopoulos and B. Drevet: Materials Science and Engineering A249(1998) 176. [Non-Patent Document 3] P. Ni, H. Goto, M. Nakamoto and T. Tanaka, ISIJ Int., Vol. 60 (2020), 1586. [Overview of the project] [Problems that the invention aims to solve]

[0006] Non-Patent Documents 1 and 2 provide qualitative explanations of the contact angle when molten metal is present on a solid surface, but they do not allow for the prediction of the contact angle. Furthermore, Non-Patent Document 3 proposes a predictive model for the contact angle in molten metal / oxide systems, but generating the model requires a lot of information about the physical properties of both the molten metal and the oxide, making it difficult to predict the contact angle in systems where information is insufficient.

[0007] This invention has been made in view of the above circumstances, and aims to provide a technology that makes it possible to easily predict the contact angle between a solid surface and molten metal. [Means for solving the problem]

[0008] As a result of diligent research, the inventors of this invention discovered that the above objective can be achieved by adopting the following configuration, and thus completed the present invention. In other words, the present invention provides the following [1] to [6].

[0009] [1] A contact angle prediction method comprising: a processor performing a process to acquire analytical information for each combination of solid and liquid, including the contact angle of the liquid on the surface of the solid and numerical information relating to the solid and the liquid; and a process to acquire a prediction model that outputs a predicted value of the contact angle by inputting the numerical information acquired for each combination, wherein the numerical information includes information on the melting point and information on thermodynamic parameters, the melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid; and the thermodynamic parameters include, if the solid is a compound, the thermodynamic parameters of the solid; if the liquid is a compound, the thermodynamic parameters of the liquid and the thermodynamic parameters of a stable compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

[0010] [2] The contact angle prediction method according to [1], wherein the solid is an oxide and the liquid is a molten metal.

[0011] [3] The contact angle prediction method according to [1] or [2], wherein the thermodynamic parameter includes one or more selected from the group consisting of formation energy, formation enthalpy, and formation free energy.

[0012] [4] A contact angle prediction device equipped with a processor, wherein the processor performs the following for each combination of solid and liquid: a process to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid and numerical information relating to the solid and the liquid; and a process to acquire a prediction model that outputs a predicted value of the contact angle by inputting the numerical information acquired for each combination, wherein the numerical information includes melting point information and thermodynamic parameter information, the melting point includes the melting point of individual elements constituting the solid, the melting point of individual elements constituting the liquid, the melting point of the solid, and the melting point of the liquid; and the thermodynamic parameter includes, if the solid is a compound, the thermodynamic parameter of the solid; if the liquid is a compound, the thermodynamic parameter of the liquid and the thermodynamic parameter of a stable compound among a compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

[0013] [5] A method for generating a contact angle prediction model, wherein the processor performs the following steps for each combination of solid and liquid: a process to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid, and information on the melting point and thermodynamic parameters relating to the solid and the liquid; and a process to generate a prediction model that outputs a predicted value of the contact angle by performing an analysis with the information on the melting point and thermodynamic parameters obtained for each combination as explanatory variables and the contact angle as the objective variable, wherein the melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid; and the thermodynamic parameters include, if the solid is a compound, the thermodynamic parameters of the solid; if the liquid is a compound, the thermodynamic parameters of the liquid, and the thermodynamic parameters of a stable compound among a compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

[0014] [6] A contact angle prediction model generation device equipped with a processor, wherein the processor performs the following processes for each combination of solid and liquid: a process to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid, and information on the melting point and thermodynamic parameters relating to the solid and the liquid; and a process to generate a prediction model that outputs a predicted value of the contact angle by performing an analysis with the information on the melting point and thermodynamic parameters obtained for each combination as explanatory variables and the contact angle as the objective variable, wherein the melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid; and the thermodynamic parameters include, if the solid is a compound, the thermodynamic parameters of the solid; if the liquid is a compound, the thermodynamic parameters of the liquid, and the thermodynamic parameters of a stable compound among a compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid. [Effects of the Invention]

[0015] According to the present invention, it becomes possible to easily predict the contact angle between a solid surface and molten metal. [Brief explanation of the drawing]

[0016] [Figure 1] This figure shows an example of a system configuration including a contact angle prediction device according to this embodiment. [Figure 2] This figure shows an example of the hardware configuration of the contact angle prediction device according to this embodiment. [Figure 3] This figure shows an example of the configuration of the dataset DB according to this embodiment. [Figure 4] This figure shows an example of the functional configuration of the contact angle prediction device according to this embodiment. [Figure 5] This figure shows an example of a flow chart for the contact angle prediction method according to this embodiment. [Figure 6] This figure shows an example of output according to this embodiment. [Figure 7] It is a diagram showing an example of the formation energy according to the composition ratio of the "Pb-Mg" system according to this embodiment. [Figure 8] It is a diagram showing an example of the formation energy according to the composition ratio of the "Pb-O" system according to this embodiment. [Figure 9] It is a diagram showing an example of the formation energy according to the composition ratio of the "Fe-Mn" system according to this embodiment.

Mode for Carrying Out the Invention

[0017] [Findings Obtained by the Present Inventors]<000098>The present inventors obtained the following findings.

[0018] In Non-Patent Document 3, in predicting the contact angle of a molten metal on an oxide, various physical property values regarding the molten metal and the oxide are required. The reason is considered to be based on the idea that the wetting and spreading behavior of the molten metal on the oxide is determined mainly by the physical property values of the oxide and the molten metal respectively, and if machine learning is carried out using those physical property values, the contact angle can be predicted.

[0019] Therefore, the inventors earnestly studied what physical property values to use, and conceived using physical property values based on the idea that the phase stability of substances composed of the elements constituting the solid and the liquid affects the interfacial properties between the solid and the liquid (solid / liquid) that forms a contact angle on the solid.

[0020] In other words, by comparing the case where a solid exists in a solid state with the case where a solid reacts with elements constituting a liquid to form a new compound, we can discuss whether the solid exists stably or whether it reacts with elements constituting a liquid to form a new, different solid. We conceived that the contact angle can be estimated from this reactivity between the solid and the liquid. Furthermore, we found that in order to achieve this, thermodynamic data and melting points of compounds and pure substances obtained from all combinations of elements constituting the solid-liquid system to be evaluated are necessary. We found that the contact angle can be predicted using these, and thus the present invention was completed. This invention is based on the above findings.

[0021] Next, preferred embodiments of the present invention will be described. For convenience of explanation, the following description may sometimes be based on the perspective of a GUI (Graphical User Interface). Furthermore, since the fundamental data processing technologies for realizing the present invention (communication / transmission technologies, data acquisition technologies, data recording technologies, data processing / analysis technologies, machine learning technologies, image processing technologies, and visualization technologies, etc.) are known technologies, their explanation will be omitted.

[0022] Furthermore, in this specification, the concept of "device" includes not only a single device that performs a specific function, but also a combination of multiple devices that exist independently and in a distributed manner while cooperating (linking) to perform a specific function.

[0023] Furthermore, in this specification, "user" refers to a person who conducts research and development regarding the interfacial properties between a solid and a liquid that forms a contact angle on the solid (solid / liquid). Specifically, it refers to a person who, for example, enjoys information and services related to such research and development through the functions of the contact angle prediction device, etc., of the present invention. However, this definition is merely an example adopted for the sake of simplifying the explanation and does not limit the forms to which the invention can be applied to such a definition of user.

[0024] Furthermore, in this specification, "person," "entity," or "user" refers to a natural person performing a specific action, and may include individuals or groups. Also, in this specification, the contact angle prediction model generated by and used by the contact angle prediction device may include not only artificial intelligence (AI) such as learning models obtained through deep learning, but also regression equations obtained by multiple regression analysis based on the explanatory and dependent variables in the generation of the learning model. Artificial intelligence (AI) realizes intelligent functions such as reasoning, prediction, and judgment using hardware and software resources. The algorithm of the artificial intelligence is arbitrary and may include, for example, expert systems, case-based reasoning (CBR), Bayesian networks, or inclusion architectures.

[0025] [Contact angle prediction method] The contact angle prediction method of this embodiment involves a processor performing the following steps: acquiring analytical information for each combination of solid and liquid, including the contact angle of the liquid on the surface of the solid and numerical information relating to the solid and the liquid; and analyzing the analytical information acquired for each combination to acquire a prediction model that outputs a predicted value of the contact angle when the numerical information is input. The numerical information includes melting point information and thermodynamic parameter information. The melting point information includes the melting points of the elements constituting the solid, the melting points of the elements constituting the liquid, the melting point of the solid, and the melting point of the liquid. The thermodynamic parameter information includes the thermodynamic parameters of compounds composed of at least one of the elements constituting the solid and the elements constituting the liquid, which satisfy the set conditions regarding energy stability. The following description also serves as a description of the method for generating the contact angle prediction model.

[0026] [Various information acquisition processes] In the contact angle prediction method of this embodiment, analytical information is acquired for each combination of solid and liquid, including the contact angle of the liquid on the surface of the solid, and numerical information relating to the solid and the liquid. First, the solids and liquids for which information is to be acquired are shown below.

[0027] (solid) A solid can be any substance in a solid state, and can be either a pure substance or a mixture. Pure substances include elements composed of only one type of element and compounds composed of two or more elements. Examples of elements include metals. Examples of compounds include alloys classified as intermetallic compounds and inorganic compounds (oxides, nitrides, carbides, etc.). Examples of mixtures include alloys classified as substitutional solid solutions and interstitial solid solutions. The solid is preferably an oxide. When the solid is an oxide, there is a large amount of data on measured contact angles, which allows for the construction of a highly accurate model. For solids, if the impurity content is 5% by mass or less, the error range of the predicted contact angle will be 5% or less. Therefore, the impurity content of solids can be tolerated up to 5% by mass.

[0028] (liquid) The liquid can be any substance in a liquid state and is not particularly limited. However, systems in which the atoms constituting the liquid have strong interactions, specifically substances that form a network structure even in a liquid state, such as SiO4, or substances with strong hydrogen bonding, such as water, are undesirable. On the other hand, molten metals or molten metalloids are preferred because, in the liquid state, they do not easily undergo local ordering and can be described by the macroscopic physical properties of the liquid itself. For liquids containing 5% by mass or less of impurities, the error range of the predicted contact angle will be 5% or less. Therefore, an impurity content of 5% by mass or less in the liquid is acceptable.

[0029] (contact angle) The contact angle is the contact angle between the liquid and the solid surface. The contact angle may be a measured value or a value from literature. For measuring the contact angle, for example, a value measured using in-situ high-temperature observation may be used.

[0030] (Numerical information: melting point) The numerical information in this embodiment includes information on the melting point and information on thermodynamic parameters. Of these, the melting point is an index that allows for a relative comparison of the stability of the liquid state. Here, the melting point includes the melting points of all elemental elements selected from the elements constituting the solid and the liquid, and the melting points of the solid and the liquid. Here, elements contained in the solid and the liquid at a concentration of 1 mass or less do not need to be considered.

[0031] Specifically, if compound A is a solid composed of elements α, β, and γ, and compound B is a liquid composed of elements η and ε, then these are the melting points of element α, element β, element γ, element η, element ε, compound A, and compound B, respectively.

[0032] The melting point is a parameter that reflects the relative stability between the solid and liquid states of a substance, and can be used to evaluate the relative stability between the solid and liquid states. Here, the melting point may be calculated using a phase diagram, or it may be estimated by proportionally distributing the melting points of each element.

[0033] (Numerical information: Thermodynamic parameters) If the solid is a compound (hereinafter also referred to as "compound A" for convenience), the thermodynamic parameters of the solid (compound A) are used. Furthermore, if the liquid is a compound (hereinafter referred to as "compound B" for convenience), the thermodynamic parameters of the liquid (compound B) are used. Furthermore, it is important to use the thermodynamic parameters of a stable compound (hereinafter, also referred to as "compound C" for convenience) composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid, as thermodynamic parameters. Compound C is a compound of all combinations of the two elements (one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid). However, elements present in solids and liquids at a concentration of 1% by mass or less do not need to be considered, as their thermodynamic effect on the contact angle is not significant.

[0034] Here, a stable compound refers to a compound C whose reported thermodynamic parameters (such as energy of formation) are relatively small. For example, a stable compound is one whose thermodynamic parameters fall within ±20% of the thermodynamic parameters of the most stable compound (e.g., the compound with the lowest values ​​of thermodynamic parameters such as energy of formation).

[0035] The stable compound is not limited to one, but may be two or more. That is, the thermodynamic parameters of two or more stable compounds may be used. Specifically, for example, from the energy curve shown in Figure 8 later, thermodynamic parameters (formation energy) may be adopted not only for PbO, the compound with the lowest value, but also for Pb3O4, the second lowest compound, and PbO2, the third lowest compound. In this case, information such as the fact that PbO has the lowest formation energy; that stable compounds exist on the side with more oxygen than PbO; that compounds with more oxygen, such as PbO2, have higher formation energies; and that no stable compounds exist on the side with more Pb than PbO can be incorporated. This is expected to improve the accuracy of the resulting contact angle prediction model.

[0036] Since the contact angle is determined from the relative stability of the solid and liquid states of a substance, it is preferable to use at least one thermodynamic parameter selected from the group consisting of formation energy, formation enthalpy, and formation free energy. For improved accuracy, it is also preferable to use either the constant-pressure specific heat or a physical property value correlated with the constant-pressure specific heat (e.g., bulk modulus), or both, as thermodynamic parameters. The bulk modulus has the advantage of being readily available as it can be obtained relatively easily from first-principles calculations, and therefore it is preferable to use the bulk modulus as well.

[0037] Thermodynamic parameters may be values ​​already recorded in publicly available databases, such as the NIST-JANAF database, or thermodynamic parameters obtained from first-principles calculations, such as those in the Materialsproject database, or values ​​calculated using any first-principles calculation, molecular dynamics calculation, or thermodynamic calculation.

[0038] Specifically, if compound A is composed of elements α, β, and γ as a solid, and compound B is composed of elements η and ε as a liquid, then the thermodynamic parameters of the solid (compound A) and the liquid (compound B) are used. In addition, at least one thermodynamic parameter of a stable compound C, which is composed of one element selected from the elements (α, β, γ) constituting compound A and one element selected from the elements (η, ε) constituting compound B, is used. As an example, Figures 7 and 8 show the relationship between the composition ratio of elements and the energy of formation, which are thermodynamic parameters for stable compounds such as the elements "Pb-Mg" and "Pb-O". In the example shown in Figure 7, for example, Mg2Pb, which has the lowest energy of formation, is adopted as the stable compound. Similarly, in the example shown in Figure 8, for example, PbO, which has the lowest energy of formation, is adopted as the stable compound.

[0039] Specifically, the thermodynamic parameters of compound A, composed of elements α, β, and γ, and the thermodynamic parameters of compound B, composed of elements η and ε, are used, along with the thermodynamic parameters of at least one compound from among those stable in the α-η system, α-ε system, β-η system, β-ε system, γ-η system, and γ-ε system.

[0040] However, in the case of a system of two elements for which no stable compounds exist (see Figure 9), for example, an alloy system without intermetallic compounds, such as an Fe-Mn alloy, one should select the compound with the most stable composition reported in the first-principles calculation database. If no such value is available, one should use a value that corresponds to an enthalpy of formation of a compound with an elemental ratio of 1:1 of +30 kJ / mol.

[0041] [Generating a Contact Angle Prediction Model] In generating the contact angle prediction model in this embodiment, an analysis is performed using the melting point and thermodynamic parameters, which are analytical information, as explanatory variables, and the contact angle as the objective variable to generate a prediction model that outputs a predicted value of the contact angle. In other words, the generation of the contact angle prediction model is done by using a dataset of objective variables and explanatory variables whose correspondence is known in advance.

[0042] Multiple regression analysis, machine learning, or deep learning can be used to generate a contact angle prediction model. When generating a contact angle prediction model, it is preferable to select a target variable that is similar to the contact angle to be predicted, for which the correspondence is known in advance. Specifically, if the contact angles used in the training data are for solid oxides, it will be a suitable model for predicting the contact angle of a liquid on a solid oxide, and if the contact angles are for molten metals, it will be a suitable model for predicting the contact angle of molten metals on a solid.

[0043] When applying the above machine learning or deep learning methods, it is preferable to generate a contact angle prediction model using, for example, five or more datasets. For machine learning, a random forest model can be used, and for deep learning, known methods such as U-Net can be employed. The specific configurations of the contact angle prediction model generation device and the contact angle prediction device, which generate such contact angle prediction models and use them to predict contact angles, will be described later.

[0044] [Contact angle prediction device] The effects of this embodiment will be described in detail below based on examples. However, this embodiment is not limited to these examples. Here, the specific configuration of the contact angle prediction device in this embodiment will be illustrated with reference to Figure 1. Figure 1 is a diagram showing an example of a system configuration including the contact angle prediction device 10 according to this embodiment. The contact angle prediction device 10 is connected to a management terminal 20 so as to be able to communicate via an appropriate network N or via an internal BUS.

[0045] The management terminal 20 described above is a terminal used by users who perform contact angle prediction using the contact angle prediction device 10. Specifically, this management terminal 20 consists of a PC (Personal Computer), a notebook PC, a smartphone, a tablet terminal, etc. The management terminal 20 delivers a dataset for generating a contact angle prediction model to the contact angle prediction device 10 in response to user instructions or requests from the contact angle prediction device 10. It also acquires the contact angle prediction results from the contact angle prediction device 10 and outputs them to a display or the like for the user to view.

[0046] On the other hand, the contact angle prediction device 10 is specifically composed of a server device, a PC (Personal Computer), or a notebook PC, etc. As described above, it generates a contact angle prediction model 122 from a dataset provided by the management terminal 20 or input from its own input device 14, and performs contact angle prediction processing using this model.

[0047] As shown in Figure 2, the computer comprising the contact angle prediction device 10 includes a processor 11, an auxiliary storage device 12, a main storage device 13, an input device 14, an output device 15, and a communication device 16. The processor 11 is composed of, for example, a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), an MCU (Micro Controller Unit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), a TPU (Tensor Processing Unit), or an ASIC (Application Specific Integrated Circuit).

[0048] Furthermore, the auxiliary storage device 12 is composed of, for example, flash memory, HDD (Hard Disc Drive), SSD (Solid State Drive), FD (Flexible Disc), MO disk (Magneto-Optical disc), CD (Compact Disc), DVD (Digital Versatile Disc), SD card (Secure Digital card), or USB memory (Universal Serial Bus memory).

[0049] Such auxiliary storage devices 12 may be built into the computer main body that constitutes the contact angle prediction device 10, or they may be attached to the computer main body as external devices. Alternatively, the auxiliary storage device 12 may be configured as a NAS (Network Attached Storage) or the like. Furthermore, the auxiliary storage device 12 may be an external device that can communicate with one of the computers that constitutes the contact angle prediction device 10 via a communication network, such as an online storage or database server.

[0050] Furthermore, the auxiliary storage device 12 holds the operating system (OS) and a program 121 that includes a contact angle prediction model, i.e., the contact angle prediction model 122. The contact angle prediction model 122 is a model that takes numerical information (information on melting point and thermodynamic parameters) input from the input device 14 or management terminal 20 as input and outputs information on the contact angle of the liquid with respect to the target solid.

[0051] The above contact angle prediction model 122 may employ artificial intelligence (AI), such as a learning model obtained through deep learning. In this case, the contact angle prediction model 122 is a learning model obtained by performing deep learning on a dataset (training data) in which the above numerical information and numerical contact angles are paired. When the above program 121 is read and executed by the processor 11, the computer constituting the contact angle prediction device 10 performs the functions of generating the contact angle prediction model 122 and performing contact angle prediction using the contact angle prediction model 122. These functions will be described later.

[0052] In addition to the program 121 and the contact angle prediction model 122, the auxiliary storage device 12 also stores at least the data set DB 123. This data set DB 123 has the configuration shown in Figure 3, for example. The data set DB 123 is a database in which the contact angle prediction device 10 stores information consisting of numerical information in this embodiment, namely the melting points and thermodynamic parameters of the solid and liquid, and the contact angle of the liquid with respect to the solid. As shown in Figure 3, this data set DB 123 is a collection of records containing values ​​such as combinations, contact angles, elements, melting points, formation energy, and bulk modulus, with an ID that uniquely identifies the data set as the key.

[0053] Of these, "combination" indicates a pattern of solid and liquid combinations. "Contact angle" indicates the contact angle in this pattern. "Element" indicates the element included in the above combination. "Melting point" indicates the melting point of each element. "Formation energy" indicates the formation energy for the solid and each combination of elements in the above "combination". "Bulk modulus" indicates the bulk modulus for the solid and each combination of elements in the above "combination".

[0054] Now, let's return to the explanation based on Figure 2. The main memory 13 shown in Figure 2 is composed of semiconductor memory such as ROM (Read Only Memory) and RAM (Random Access Memory). The processor 11 loads the above program 121 onto the main memory 13 and executes it there.

[0055] The input device 14 is a device that accepts user input and consists of, for example, a keyboard, mouse, touch panel, microphone, etc. The output device 15 consists of, for example, a display, speaker, etc.

[0056] The communication device 16 may be composed of, for example, a network interface card or a communication interface board. The computer constituting the contact angle prediction device 10 can communicate with other devices such as a management terminal 20 connected to a network N such as the Internet and a mobile communication line via the communication device 16.

[0057] (Functional part) In this embodiment, the contact angle prediction device 10 implements the functional unit shown in Figure 4 when the processor 11 executes the program 121 described above. Figure 4 is a diagram showing an example of the functional configuration of the contact angle prediction device 10 according to this embodiment. This functional unit consists of a data acquisition unit 110, a processing unit 111, and a prediction result output unit 112.

[0058] Of these, the data acquisition unit 110 is a functional unit that receives data entered by the user from the management terminal 20 or input device 14. This data includes a dataset used to generate the contact angle prediction model 122, as well as numerical information (melting point and thermodynamic parameters) to be added to the contact angle prediction model 122 in order to obtain a predicted value of the contact angle.

[0059] Furthermore, the processing unit 111 consists of a model generation unit 1111 and an estimation unit 1112. The model generation unit 1111 is a functional unit that generates a contact angle prediction model 122 by, for example, providing the dataset obtained by the data acquisition unit 110 to a predetermined learning engine. More specifically, a form can be adopted in which a certain number or more of the above datasets are provided as supervised data to a deep learning model, and the model learns the correspondence between numerical information and contact angles. However, instead of AI obtained by such deep learning, known tools such as a solver that calculates a regression equation, which is a mathematical model, by multiple regression analysis may be appropriately adopted.

[0060] Furthermore, the estimation unit 1112 is a functional unit that adds numerical information regarding the combination of solid and liquid for which the user wishes to predict the contact angle to the contact angle prediction model 122, and obtains the contact angle value as an output. The numerical information added to the contact angle prediction model 122 is obtained from the data acquisition unit 110 via the management terminal 20 or input device 14.

[0061] Furthermore, the prediction result output unit 112 is a functional unit that outputs the contact angle prediction result, which is the processing result of the estimation unit 1112, to the output device 15 or the management terminal 20.

[0062] The contact angle prediction method in this embodiment will be described in detail below with reference to Figure 5. Figure 5 is a diagram showing an example of the flow chart of the contact angle prediction method according to this embodiment. Note that the examples of the contact angle prediction method described below are merely examples and are not limited to these examples.

[0063] (Example 1) First, the management terminal 20 (or input device 14) distributes the dataset to the contact angle prediction device 10 via the network N (S1). This dataset is the data necessary for generating the contact angle prediction model 122 described earlier, and includes the contact angle, element type and their respective melting points, formation energy, and bulk modulus for each combination of solid and liquid. The contact angle prediction device 10 may also access external databases such as the NIST database or the MaterialsProject database via the network N and obtain information corresponding to the dataset from those databases.

[0064] Meanwhile, the contact angle prediction device 10 receives the above dataset and stores it in the dataset DB123 of the auxiliary storage device 12 (S2). The contact angles that constitute this dataset correspond to the target variable. In this embodiment, all the contact angle values ​​described in Fig. 2 of Non-Patent Literature 2 are used as such contact angles. The numerical information that constitutes the above dataset corresponds to the explanatory variables. In this embodiment, this numerical information includes at least the melting point and the formation energy of the solid (compound) for combinations of solids and liquids for which the contact angle values ​​are known. The melting point is a value obtained from the NIST database, and the formation energy is a value obtained from the MaterialsProject database.

[0065] Specific examples of the values ​​that make up the above dataset will be explained using, for example, the case of liquid: Sn and solid: Fe3O4. For the melting points, the melting points of metallic Sn, metallic Fe, elemental O2, and compound Fe3O4 were used. In addition, for the thermodynamic parameters, in addition to the free energy of formation of the compound solid (Fe3O4), the free energy of formation of compound FeSn, which is stable in the Fe-Sn system, and compound SnO2, which is stable in the Sn-O system were used (note that the liquid was an element and not a compound, so the free energy of formation of the liquid was not used). Furthermore, an investigation was also conducted by adding bulk modulus to the thermodynamic parameters (already shown in Figure 3). For the bulk modulus, the bulk modulus of compounds FeSn, Fe3O4, and SnO2 obtained from the MaterialsProject database was used.

[0066] Next, the contact angle prediction device 10 performs multiple regression analysis on the dataset obtained in step S2, for example, as a machine learning technique, to obtain a contact angle prediction model 122 using multiple regression (S3). In step S3, the contact angle prediction device 10 employs a neural network type model for deep learning, and learns the model by assigning it the dataset mentioned above to obtain the contact angle prediction model 122. The contact angle prediction model 122 obtained by multiple regression analysis is referred to as "contact angle prediction model 1". The contact angle prediction model 122 obtained by deep learning is referred to as "contact angle prediction model 2".

[0067] Next, the contact angle prediction device 10 receives numerical information regarding the combination of solid and liquid to be predicted from the management terminal 20 or input device 14 as a prediction request (S4). Here, for example, suppose a contact angle prediction request is received for a case where the solid is at least one of MnO, Mn2SiO4, MnSiO3, and SiO2, and the liquid is molten zinc (Zn).

[0068] In this embodiment, as part of the numerical information accepted as a prediction request, when the solid and liquid combination is liquid:Zn and solid:MnO, the melting points of metallic Zn, metallic Mn, elemental O2, and compound MnO were adopted. Furthermore, as part of the thermodynamic parameters in the numerical information, for example, the free energy of formation of the compound solid (MnO), the free energy of formation of the compound MnZn3 which is stable in the Mn-Zn system, and the compound ZnO which is stable in the Zn-O system were adopted. In addition, the bulk moduli of compounds MnZn3, MnO, and ZnO obtained from the MaterialsProject database were adopted as thermodynamic parameters.

[0069] Furthermore, in the case of a solid-liquid combination of liquid:Zn and solid:Mn2SiO4, the melting point information was taken from metallic Zn, metallic Mn, elemental Si, elemental O2, and compound Mn2SiO4. In addition, the thermodynamic parameters were taken from the free energy of formation of the solid compound (Mn2SiO4), the compound MnZn3 which is stable in the Mn-Zn system, the compound SiZn3 which is not stable in the Si-Zn system but is reported in the MaterialsProject database, the compound MnSi which is stable in the Mn-Si system, the compound ZnO which is stable in the Zn-O system, the compound MnO which is stable in the Mn-O system, and the compound SiO2 which is stable in the Si-O system. Furthermore, the bulk moduli of compounds MnZn3, SiZn3, MnSi, ZnO, MnO, and SiO2 obtained from the MaterialsProject database were used as thermodynamic parameters.

[0070] Furthermore, in the case where the solid-liquid combination is liquid:Zn and solid:MnSiO3, the melting point information was taken from metallic Zn, metallic Mn, elemental Si, elemental O2, and compound MnSiO3. In addition, the thermodynamic parameters were taken from the free energy of formation of the compound (solid MnSiO3), the compound MnZn3 which is stable in the Mn-Zn system, the compound SiZn3 which is not stable in the Si-Zn system but is reported in the MaterialsProject database, the compound MnSi which is stable in the Mn-Si system, the compound ZnO which is stable in the Zn-O system, the compound MnO which is stable in the Mn-O system, and the compound SiO2 which is stable in the Si-O system. Furthermore, the bulk moduli of compounds MnZn3, SiZn3, MnSi, ZnO, MnO, and SiO2 obtained from the MaterialsProject database were used as thermodynamic parameters.

[0071] Furthermore, in the case where the solid-liquid combination is liquid:Zn and solid:SiO2, the melting point information was taken from metallic Zn, elemental Si, elemental O2, and compound SiO2. In addition, for thermodynamic parameters, the free energy of formation of the compound (SiO2), the compound SiZn3 (which is not stable in the Si-Zn system but is reported in the MaterialsProject database), and the compound ZnO (which is stable in the Zn-O system) were taken. Furthermore, the bulk moduli of the compounds SiZn3, MnO, and ZnO obtained from the MaterialsProject database were used as thermodynamic parameters.

[0072] The contact angle prediction device 10 assigns numerical information indicated by the prediction request received in S4 to each of the contact angle prediction models 1 and 2 generated in S3, and performs contact angle prediction between the target solid and liquid (S5). The contact angle prediction device 10 also outputs the contact angle prediction result obtained in S5 (see screen G1 in Figure 6) to the management terminal 20 or output device 15 (S6), and this flow ends.

[0073] Table 1 compares the contact angles obtained from the above-mentioned contact angle prediction model 122 (contact angle prediction models 1 and 2) with experimental values ​​for the given solid and liquid combination. The experimental values ​​are those described in existing literature (Y. Kim, M. Shin, C. Tang and J. Lee: Metallurgical and Materials Transactions B, 41(2010) 872.), and the values ​​used were for a molten metal composition of Zn-0.12 mass%Al.

[0074] [Table 1]

[0075] As can be seen from Table 1, by using either "Contact Angle Prediction Model 1" or "Contact Angle Prediction Model 2," it became possible to predict contact angle values ​​that match experimental values ​​with an accuracy of approximately ±20°. Furthermore, it was found that adding the bulk modulus to the thermodynamic parameters improved the prediction accuracy.

[0076] (Example 2) Next, the contact angle is predicted using "Contact Angle Prediction Model 1" and "Contact Angle Prediction Model 2" used in Example 1, when the solid is Al2O3 and the liquid is at least one of Fe-50%Co alloy, Fe-50%Ni alloy, or Ni-50%Co alloy by mass%. That is, the contact angle prediction device 10 receives a request from the management terminal 20 or input device 14 to predict the contact angle when the solid is Al2O3 and the liquid is at least one of Fe-50%Co alloy, Fe-50%Ni alloy, or Ni-50%Co alloy by mass% (S4).

[0077] When the combination of solid and liquid targeted by the prediction request is liquid: Fe-50%Co and solid: Al2O3, the melting point information accepted as numerical information for the prediction request includes the melting points of metallic Al, metallic Fe, metallic Co, elemental O2, and compound Al2O3, as well as the melting point of Fe-50%Co in mass percent. Furthermore, the thermodynamic parameters of the above numerical information include the free energy of formation of the compound (solid Al2O3), the free energy of formation of compounds stable in the Fe-Al system (Al2Fe), stable in the Co-Al system (AlCo), stable in the Fe-Co system (FeCo), stable in the Fe-O system (Fe2O3), and stable in the Co-O system (CoO). In addition, the bulk modulus of compounds Al2O3, Al2Fe, AlCo, FeCo, Fe2O3, and CoO obtained from the MaterialsProject database is used as a thermodynamic parameter.

[0078] Furthermore, when the combination of solid and liquid targeted for prediction is liquid: Fe-50%Ni, solid: Al2O3, the melting point information used includes the melting points of metallic Al, metallic Fe, metallic Ni, elemental O2, and compound Al2O3, as well as the melting point information of Fe-50%Ni in mass percent. The thermodynamic parameters used include the free energy of formation of the compound (solid Al2O3), the free energy of formation of the compound Al2Fe (stable in the Fe-Al system), the compound NiAl (stable in the Ni-Al system), the compound FeNi3 (stable in the Fe-Ni system), the compound Fe2O3 (stable in the Fe-O system), and the compound NiO (stable in the Ni-O system). In addition, the bulk moduli of the compounds Al2O3, Al2Fe, NiAl, FeNi3, Fe2O3, and NiO, obtained from the MaterialsProject database, were used as thermodynamic parameters.

[0079] Furthermore, when the combination of solid and liquid targeted for prediction is liquid: Ni-50%Co and solid: Al2O3, the melting points used were the melting points of metallic Al, metallic Co, metallic Ni, elemental O2, and compound Al2O3, as well as the melting point of Ni-50%Co in mass percent. The thermodynamic parameters used were the free energy of formation of the compound (solid Al2O3), the free energy of formation of the compound AlCo (stable in the Co-Al system), the compound NiAl (stable in the Ni-Al system), the compound Co3Ni (stable in the Co-Ni system), the compound CoO (stable in the Co-O system), and the compound NiO (stable in the Ni-O system). Additionally, the bulk moduli of the compounds Al2O3, AlCo, NiAl, Co3Ni, CoO, and NiO, obtained from the MaterialsProject database, were used as thermodynamic parameters.

[0080] Subsequently, similar to Example 1, the contact angle prediction device 10 assigns numerical information indicated by the prediction request received in S4 to each of the contact angle prediction models 1 and 2 generated in S3, respectively, and performs contact angle prediction between the target solid and liquid (S5). The contact angle prediction device 10 also outputs the contact angle prediction result obtained in S5 (see screen G1 in Figure 6) to the management terminal 20 or output device 15 (S6), and this flow ends.

[0081] Table 2 compares the contact angles obtained from the above-mentioned contact angle prediction model 122 (contact angle prediction models 1 and 2) with the experimental values ​​for the given solid and liquid combination. The experimental values ​​were read from graphs described in existing literature (Kazumi Ogino, Hitoshi Taimatsu, Fumitada Nakatani: Journal of the Japan Institute of Metals, Vol. 46 (1982) 1049).

[0082] [Table 2]

[0083] (Example 3) Furthermore, the "contact angle prediction model 1" and "contact angle prediction model 2" used in Example 1 are used to predict the contact angle when the solid is Fe and the liquid is an FeO-16.1%SiO2 alloy in mass%. That is, the contact angle prediction device 10 receives a request from the management terminal 20 or input device 14 to predict the contact angle when the solid is Fe and the liquid is an FeO-16.1%SiO2 alloy in mass% (S4).

[0084] For the combination of solid and liquid targeted in the prediction request, where liquid is FeO-16.1%SiO2 and solid is Fe, the melting point information used included the melting points of metallic Fe, elemental Si, and elemental O2, as well as the melting point of FeO-16.1%Co in mass percent. Furthermore, the thermodynamic parameters used included the free energy of formation of the solid (Fe), the free energy of formation of the stable compound FeO in the Fe-O system, the stable compound SiO2 in the Si-O system, and the stable compound FeSi in the Fe-Si system. Additionally, the bulk moduli of the compounds FeO, SiO2, and FeSi, obtained from the MaterialsProject database, were used as thermodynamic parameters.

[0085] Subsequently, similar to Example 1, the contact angle prediction device 10 assigns numerical information indicated by the prediction request received in S4 to each of the contact angle prediction models 1 and 2 generated in S3, respectively, and performs contact angle prediction between the target solid and liquid (S5). The contact angle prediction device 10 also outputs the contact angle prediction result obtained in S5 (see screen G1 in Figure 6) to the management terminal 20 or output device 15 (S6), and this flow ends.

[0086] Table 3 compares the contact angles obtained from the above-mentioned contact angle prediction model 122 (contact angle prediction models 1 and 2) with experimental values ​​for the given solid and liquid combination. The experimental values ​​were read from a graph described in existing literature (Hanao Masafumi: Iron and Steel, Vol. 105 (2019) 595.) where the solid was an ultra-low carbon steel sheet.

[0087] [Table 3] [Explanation of Symbols]

[0088] N Network 1. Contact Angle Prediction System 10. Contact Angle Prediction Device (Contact Angle Prediction Model Generation Device) 11 processors 110 Data Acquisition Unit 111 Processing Unit 1111 Model Generation Unit 1112 Estimation Department 112 Prediction Result Output Unit 12 Auxiliary storage 121 Programs 122 Contact Angle Prediction Model 123 Dataset DB 13 Main memory 14 Input devices 15 Output device 16. Communication equipment 20 Management terminals

Claims

1. The processor, For each combination of solid and liquid, a process is performed to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid and numerical information relating to the solid and the liquid. A process to obtain a prediction model that outputs a predicted value of the contact angle by inputting the numerical information obtained for each of the above combinations, Execute, The aforementioned numerical information includes information on the melting point and information on thermodynamic parameters. The melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid. The aforementioned thermodynamic parameters are: If the solid is a compound, the thermodynamic parameters of the solid and If the liquid is a compound, the thermodynamic parameters of the liquid and A method for predicting a contact angle, comprising: the thermodynamic parameters of a stable compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

2. The contact angle prediction method according to claim 1, wherein the solid is an oxide and the liquid is a molten metal.

3. The contact angle prediction method according to claim 1 or 2, wherein the thermodynamic parameter includes one or more selected from the group consisting of formation energy, formation enthalpy, and formation free energy.

4. A contact angle prediction device equipped with a processor, The aforementioned processor, For each combination of solid and liquid, a process is performed to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid and numerical information relating to the solid and the liquid. A process to obtain a prediction model that outputs a predicted value of the contact angle by inputting the numerical information obtained for each of the above combinations, Execute, The aforementioned numerical information includes information on the melting point and information on thermodynamic parameters. The melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid. The aforementioned thermodynamic parameters are: If the solid is a compound, the thermodynamic parameters of the solid and If the liquid is a compound, the thermodynamic parameters of the liquid and A contact angle prediction device comprising: thermodynamic parameters of a stable compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

5. A method for generating a contact angle prediction model, The processor, For each combination of solid and liquid, a process is performed to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid, and information on the melting point and thermodynamic parameters of the solid and the liquid. For each combination, the analysis is performed using the melting point and thermodynamic parameters as explanatory variables and the contact angle as the target variable, thereby generating a predictive model that outputs a predicted value of the contact angle. The melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid. The aforementioned thermodynamic parameters are: If the solid is a compound, the thermodynamic parameters of the solid and If the liquid is a compound, the thermodynamic parameters of the liquid and A method for generating a contact angle prediction model, comprising: thermodynamic parameters of a stable compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

6. A contact angle prediction model generation device equipped with a processor, The aforementioned processor, For each combination of solid and liquid, a process is performed to acquire analytical information for each combination, including the contact angle of the liquid on the surface of the solid, and information on the melting point and thermodynamic parameters of the solid and the liquid. A process to generate a predictive model that outputs a predicted value of the contact angle by performing an analysis using the melting point and thermodynamic parameters as explanatory variables and the contact angle as the target variable from the analysis information obtained for each combination, Execute, The melting point includes the melting point of the element constituting the solid, the melting point of the element constituting the liquid, the melting point of the solid, and the melting point of the liquid. The aforementioned thermodynamic parameters are: If the solid is a compound, the thermodynamic parameters of the solid and If the liquid is a compound, the thermodynamic parameters of the liquid and A contact angle prediction model generating apparatus, comprising: thermodynamic parameters of a stable compound composed of one element selected from the elements constituting the solid and one element selected from the elements constituting the liquid.

Images