Information processing device, dielectric loss tangent prediction method, and program

Abstract

This information processing device comprises: a calculating means that calculates the molecular surface area and the molecular volume of a compound to be predicted, the molecular structure of which is in an energetically stable state; a predicting means that uses a regression model representing a correlation of the molecular surface area and the molecular volume with the dielectric loss tangent to predict the dielectric loss tangent of the compound to be predicted; and an outputting means that outputs the predicted dielectric loss tangent.

Description

Information Processing Apparatus, Dielectric Loss Tangent Prediction Method, and Program

[0001] The present disclosure relates to an information processing apparatus, a dielectric loss tangent prediction method, and a program.

[0002] As the demand for faster and larger-capacity information transmission increases, for example, in next-generation high-speed communication (6G), it is expected to use millimeter-wave bands or sub-terahertz bands, which are higher-frequency bands than before. In high-frequency bands, dielectric loss becomes a problem. Dielectric loss is a phenomenon in which the electrical energy of an applied alternating electric field is lost as heat. In order to suppress dielectric loss, it is necessary to make the dielectric loss tangent (hereinafter sometimes referred to as Df) of the substrate material constituting the communication device smaller (lower dielectric constant).

[0003] In recent years, hydrocarbons, which have attracted attention as low-dielectric substrate resins, have the feature of extremely small Df, but also have the feature of being easily combustible due to heat generation. Therefore, it is necessary to add a flame retardant to hydrocarbons for the purpose of minimizing combustion. The flame retardant added to hydrocarbons is required to have a smaller Df itself in order not to deteriorate the dielectric properties.

[0004] Conventionally, the evaluation of the Df of a flame retardant requires man-hours for raw material procurement, synthesis, purification, preparation of samples for Df measurement, and Df measurement of the flame retardant. In Patent Document 1, a physical property prediction apparatus and method for a resin composition that predicts the elongation and tensile strength of a resin composition in which many materials such as a flame retardant, a flame retardant aid, an antioxidant, a copper damage inhibitor, a lubricant, a colorant, and a crosslinking aid are used in addition to a base polymer have been proposed (for example, see Patent Document 1).

[0005] Japanese Patent No. 7416155

[0006] For example, in an existing method for predicting the Df of a flame retardant, which is an example of a compound to be predicted, from information obtained by quantum chemical calculation and molecular dynamics method based on linear response theory, there is a problem that the experimental tendency of the Df of the flame retardant alone may not be reproduced. Also, the existing method has a problem of requiring a huge calculation cost.

[0007] This disclosure aims to provide an information processing device, a dielectric loss tangent prediction method, and a program that predict the dielectric loss tangent of a target compound with lower computational costs and higher accuracy.

[0008] This disclosure comprises the following configuration.

[0009] [1] An information processing apparatus comprising: a calculation means for calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction means for predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and the molecular volume and the dielectric loss tangent; and an output means for outputting the predicted dielectric loss tangent.

[0010] [2] The information processing apparatus according to [1], further comprising: a regression model creation means for creating the regression model using the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state as explanatory variables and the dielectric loss tangent of the target compound as the objective variable.

[0011] [3] The information processing apparatus according to [2], further comprising: molecular structure calculation means for calculating the molecular structure in which the predictable compound or the learning target compound is in the stable state.

[0012] [4] The information processing apparatus according to [3], wherein the molecular structure calculation means calculates the molecular structure that is in the stable state using quantum chemical calculations or classical force field calculations.

[0013] [5] The regression model described above is

[0014] Here, Df is the measured dielectric loss tangent, S is the molecular surface area, V is the molecular volume, and C ｉ (i = 0 to 5) are regression coefficients. An information processing device as described in any one of [1] to [4].

[0015] [6] The information processing apparatus according to any one of [1] to [5], wherein the predictable compound is a flame retardant added to a substrate resin.

[0016] [7] A dielectric loss tangent prediction method comprising: a calculation procedure for calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction procedure for predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and the molecular volume and the dielectric loss tangent; and an output procedure for outputting the predicted dielectric loss tangent.

[0017] [8] A program that causes an information processing device to execute: a calculation step of calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction step of predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and the molecular volume and the dielectric loss tangent; and an output step of outputting the predicted dielectric loss tangent.

[0018] According to this disclosure, it is possible to provide an information processing device, a dielectric loss tangent prediction method, and a program that predict the dielectric loss tangent of a target compound with lower computational cost and higher accuracy.

[0019] This is a diagram showing the configuration of an example of an information processing system according to this embodiment. This is a hardware configuration diagram of an example of a computer according to this embodiment. This is a functional configuration diagram of an example of an information processing system according to this embodiment. This is a flowchart showing an example of the processing in the preparation stage according to this embodiment. This is a compound list showing an example of a target compound for prediction used in the preparation stage according to this embodiment. This is a diagram showing an example of the correlation between the Df predicted value and the Df measured value of the training data used in the processing of step S14 and the verification data predicted in the processing of step S20. This is a flowchart showing an example of the processing in the prediction stage according to this embodiment.

[0020] Next, embodiments of the present invention will be described in detail. However, the present invention is not limited to the following embodiments.

[0021] <System Configuration> Figure 1 is a configuration diagram of an example of an information processing system 1 according to this embodiment. The information processing system 1 in Figure 1 has a server device 10 and a client terminal 12. The server device 10 and the client terminal 12 of the information processing system 1 are connected to enable data communication via a communication network 18 such as a local area network (LAN) or the Internet.

[0022] The client terminal 12 is an information processing device operated by the worker, such as a PC, tablet, or smartphone. The client terminal 12 displays a screen on its display device that accepts information input from the worker and accepts information input from the worker. The client terminal 12 also sends a processing request to the server device 10 according to the worker's input and causes the server device 10 to execute the processing. The client terminal 12 receives information on the execution result of the processing by the server device 10 and displays it on its display device for the worker to confirm.

[0023] The server device 10 is an information processing device such as a PC or workstation. The server device 10 receives processing requests from client terminals 12 and executes the processing. The server device 10 sends processing result information to the client terminals 12 and causes the client terminals 12 to display the processing result information.

[0024] The information processing system 1 may be implemented with a server device 10 that provides a program download function to a client terminal 12, and a client terminal 12 that executes the downloaded program. The client terminal 12 that executes the program may perform the process described later to predict the Df of the target compound.

[0025] The information processing system 1 may be implemented using a server device 10 with web server functionality and a client terminal 12 that runs web applications using web browser functionality. The client terminal 12 that runs web applications using web browser functionality displays the Df (dielectric loss tangent) predicted by the server device 10 with web server functionality, as described later, on a display device, for example, for the operator to confirm. Df is a numerical value that represents the degree of electrical energy loss within the dielectric.

[0026] The information processing system 1 may be implemented by a server device 10 that executes a program and a client terminal 12 working together to perform processing. The server device 10 that executes the program and the client terminal 12 work together to predict the Df of the target compound as described below, and the predicted Df of the target compound is displayed on a display device, for example, for the operator to confirm.

[0027] It should be noted that the information processing system 1 in Figure 1 is merely an example, and there are various system configurations depending on the application and purpose. For example, the server device 10 may be implemented by multiple computers that work together to perform processing, or it may be implemented as a cloud computing service. The information processing system 1 may also be implemented by a standalone computer.

[0028] <Hardware Configuration> The server device 10 and client terminal 12 in Figure 1 are implemented by a computer 500 with the hardware configuration shown in Figure 2, for example.

[0029] Figure 2 is a hardware configuration diagram of an example of a computer 500 according to this embodiment. The computer 500 in Figure 2 includes an input device 501, a display device 502, an external interface 503, RAM (Random Access Memory) 504, ROM (Read Only Memory) 505, a CPU (Central Processing Unit) 506, a communication interface 507, and an HDD (Hard Disk Drive) 508, and each is interconnected via bus B. Note that the input device 501 and the display device 502 may be connected and used via the external interface 503.

[0030] The input device 501 is a touch panel, operation keys or buttons, keyboard, or mouse used by the operator to input various signals. The display device 502 consists of a display such as a liquid crystal or organic EL that displays an image, and a speaker that outputs sound data such as voice or sound. The communication interface 507 is an interface for the computer 500 to perform data communication.

[0031] Furthermore, the HDD 508 is an example of a non-volatile storage device that stores programs and data. The programs and data stored include the OS, which is the basic software that controls the entire computer 500, and application programs that provide various functions on the OS. The computer 500 may also use a drive device that uses flash memory as a storage medium (such as a solid-state drive (SSD)) instead of or in conjunction with the HDD 508.

[0032] The external I / F 503 is an interface to an external device. An external device may be a recording medium 503a. This allows the computer 500 to read from and / or write to the recording medium 503a via the external I / F 503. Examples of recording media 503a include flexible disks, CDs, DVDs, SD memory cards, and USB memory.

[0033] ROM 505 is an example of a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. ROM 505 stores programs and data such as the BIOS, OS settings, and network settings that are executed when the computer 500 starts up. RAM 504 is an example of a volatile semiconductor memory (storage device) that temporarily holds programs and data.

[0034] The CPU 506 is an arithmetic unit that controls and implements the functions of the entire computer 500 by reading programs and data from storage devices such as ROM 505 and HDD 508 onto RAM 504 and executing processing. In this embodiment, the computer 500 can implement various functions of the server device 10 and client terminal 12 by executing programs.

[0035] The program to be installed in RAM 504 is installed, for example, by reading the program recorded on the recording medium 503a via the external I / F 503. Alternatively, the program may be installed by downloading it from the communication network 18 via the communication I / F 507.

[0036] <Functional Configuration> The functional configuration of the information processing system 1 according to this embodiment will now be described. Figure 3 is a functional configuration diagram of an example of the information processing system 1 according to this embodiment. Note that parts of the configuration diagram in Figure 3 that are not necessary for the explanation of this embodiment have been appropriately omitted.

[0037] The server device 10 of the information processing system 1 in Figure 3 has a request receiving unit 20, a response transmission unit 22, a molecular structure calculation unit 24, a calculation unit 26, a prediction unit 28, a regression model creation unit 30, an output unit 32, and a storage unit 40. The storage unit 40 stores the regression model 42 and compound data. The client terminal 12 of the information processing system 1 has an information display unit 50, an operation reception unit 52, a request transmission unit 54, and a response receiving unit 56.

[0038] The information display unit 50 displays information on the display device 502, including a screen for receiving information input from the worker and the execution results of the server device 10's processing. The operation reception unit 52 receives operations from the worker, such as information input. The request transmission unit 54 sends a processing request to the server device 10 in response to the information input from the worker. The response reception unit 56 receives a response from the server device 10 to the processing request sent by the request transmission unit 54.

[0039] The request receiving unit 20 receives a processing request from the client terminal 12. The response transmission unit 22 responds to the client terminal 12 with the result of the processing performed in accordance with the processing request. The molecular structure calculation unit 24 optimizes the molecular structure of the target compound or the target compound by calculating the molecular structure in which the target compound or the target compound is in a stable state. The molecular structure calculation unit 24 can be implemented, for example, by an existing quantum chemistry calculation program such as Gaussian, or a classical force field calculation program. The compound data in the storage unit 40 contains the information necessary for the molecular structure calculation unit 24 to calculate the molecular structure in which the target compound or the target compound is in a stable state.

[0040] The compound to be predicted is a compound for which Df is predicted by the information processing system 1 according to the present embodiment, and is, for example, a flame retardant added to a substrate resin. The compound for learning is a compound of compound data used for creating the regression model 42, and is, for example, a flame retardant added to a substrate resin. The flame retardant added to the substrate resin is, for example, a phosphorus-based flame retardant.

[0041] The calculation unit 26 calculates the molecular surface area and molecular volume of the compound to be predicted or the compound for learning whose molecular structure is in an energetically stable state. The calculation unit 26 calculates, for example, the molecular surface area and molecular volume when the atoms of the compound to be predicted or the compound for learning are spheres with van der Waals radii.

[0042] The prediction unit 28 predicts the Df of the compound to be predicted using the regression model 42 that represents the correlation between the molecular surface area and molecular volume and Df. The regression model creation unit 30 creates the regression model 42 using the molecular surface area and molecular volume of the compound for learning with the optimal molecular structure calculated by the molecular structure calculation unit 24 as explanatory variables and the Df of the compound for learning as the objective variable. The compound for learning with the optimal molecular structure is a compound for learning whose molecular structure is in an energetically stable state. The compound data in the storage unit 40 includes information on the compound for learning necessary for the regression model creation unit 30 to create the regression model 42.

[0043] The output unit 32 outputs the Df of the compound to be predicted predicted by the prediction unit 28. The output of the Df of the compound to be predicted may be displayed on the display device of the client terminal 12, or may be stored in a storage unit (for example, the storage unit 40) that stores the Df of the compound to be predicted. The storage unit that stores the Df of the compound to be predicted may be a device other than the server device 10.

[0044] The storage unit 40 stores programs and data necessary for processing executed by the server device 10 or the client terminal 12, such as the regression model 42 and compound data.

[0045] Note that the functional block diagram of FIG. 3 is an example, and it can be implemented with various configurations. For example, the storage unit 40 may be a storage device, a computer, or cloud storage that can perform data communication with the server device 10 and the client terminal 12 via, for example, a communication network 18. Further, at least a part of the processing performed by the server device 10 in FIG. 3 may be performed by the client terminal 12.

[0046] <Processing> Hereinafter, the processing of the information processing system 1 according to the present embodiment will be described by dividing it into processing in the preparation stage and processing in the prediction stage.

[0047] FIG. 4 is a flowchart showing an example of the processing in the preparation stage according to the present embodiment. FIG. 5 is a compound list showing an example of a compound to be predicted used in the preparation stage according to the present embodiment. Hereinafter, an example will be described in which, among the compound lists shown in FIG. 5, (2) tris(4-methylphenyl)phosphine oxide, (10) PQ-60, and (12) DPPO-NQ are verification target compounds, and the others are learning target compounds.

[0048] When the server device 10 receives an instruction to create a regression model 42 from an operator, it starts the processing of the flowchart shown in FIG. 4.

[0049] In step S10, the molecular structure calculation unit 24 of the server device 10 reads out the compound data of the learning target compounds from the compound list shown in FIG. 5. The molecular structure calculation unit 24 performs optimization of the molecular structure of the learning target compounds by calculating the molecular structure in which the learning target compounds are in a stable state using the read compound data.

[0050] In step S12, the calculation unit 26 calculates the molecular surface area S and molecular volume V of the learning target compound whose molecular structure was optimized in step S10. Specifically, the calculation unit 26 calculates the molecular surface area S and molecular volume V of the following compounds included in the compound list shown in Figure 5: (1) triphenylphosphine oxide, (3) tri-p-cresyl phosphate, (4) Mosaflam-310, (5) Mosaflam-346, (6) 6H-dibenzo[c,e][1,2]oxaphosphorine, 6,6′-[[1,1′-biphenyl]-4,4′-diylbis(methylene)]bis-,6,6′-dioxide, (7) PX-202, (8) bisphenol A bis(diphenyl phosphate), (9) 6H-dibenzo[C,E][1,2]oxaphosphorine, 6,6′-(1-phenyl-1,2-ethanediyl)bis,6,6′-dioxide, (11) HCA-HQ, and (13) hexaphenoxycyclotriphosphazene.

[0051] In step S14, the regression model creation unit 30 creates the following regression model 42, using the molecular surface area S and molecular volume V of the target compound with the optimal molecular structure calculated in step S12 as explanatory variables, and the measured value Df of the target compound as the objective variable.

[0052] Df is the measured value of Df. S is the molecular surface area of ​​the target compound. V is the molecular volume of the target compound. C ｉ (i = 0 to 5) are the regression coefficients.

[0053] For example, the regression model creation unit 30 regresses the 10 training data points, which represent the correlation between the molecular surface area S and molecular volume V of the target compound calculated in step S12 and the measured value of Df, into the regression model 42.

[0054] In step S16, the molecular structure calculation unit 24 of the server device 10 reads the compound data of the compound to be verified from the compound list shown in Figure 5. Using the read compound data, the molecular structure calculation unit 24 optimizes the molecular structure of the compound to be verified by calculating the molecular structure in which the compound to be verified is in a stable state.

[0055] In step S18, the calculation unit 26 calculates the molecular surface area S and molecular volume V of the compounds whose molecular structures were optimized in step S16. Specifically, the calculation unit 26 calculates the molecular surface area S and molecular volume V of (2) tris(4-methylphenyl)phosphine oxide, (10) PQ-60, and (12) DPPO-NQ, which are included in the compound list shown in Figure 5.

[0056] In step S20, the prediction unit 28 predicts the Df prediction value for the molecular surface area S and molecular volume V calculated in step S18 using the regression model 42 created in step S14. Specifically, the prediction unit 28 uses the molecular surface area S and molecular volume V of (2) tris(4-methylphenyl)phosphine oxide, (10) PQ-60, and (12) DPPO-NQ, which are included in the compound list shown in Figure 5, as verification data to predict the Df prediction value.

[0057] In step S22, the output unit 32 displays the correlation between the Df predicted value and the Df measured value of the training data used in step S14 and the verification data predicted in step S20, for example as shown in Figure 6, thereby allowing the operator to verify the regression model 42 created in step S14.

[0058] For example, the regression coefficients of the regression model 42 created as described above using the compound list shown in Figure 5 are C0 = -3.075192, C1 = 0.0061952, C2 = -0.005495, C3 = -0.000009875, C4 = 548.5, and C5 = 549.825.

[0059] Figure 7 is a flowchart showing an example of the processing in the prediction stage according to this embodiment.

[0060] When the server device 10 receives a prediction instruction for the Df of the target compound from the operator, it starts processing the flowchart shown in Figure 7.

[0061] In step S30, the molecular structure calculation unit 24 of the server device 10 reads out the compound data of the compound to be predicted. Using the read-out compound data, the molecular structure calculation unit 24 optimizes the molecular structure of the compound to be predicted by calculating the molecular structure in which the compound to be predicted will be in a stable state.

[0062] In step S32, the calculation unit 26 calculates the molecular surface area S and molecular volume V of the target compound whose molecular structure was optimized in step S30.

[0063] In step S34, the prediction unit 28 predicts the Df prediction value for the molecular surface area S and molecular volume V of the target compound calculated in step S32, using the regression model 42 created in step S14 of Figure 4.

[0064] In step S36, the output unit 32 displays the measured Df value of the target compound predicted in the processing of step S34 on, for example, the information display unit 50 of the client terminal 12, allowing the operator to confirm it.

[0065] The information processing system 1 according to this embodiment can predict the Df of a target compound with less computational cost than existing methods, and can also predict the Df of a single target compound with greater accuracy than existing methods.

[0066] As described above, the information processing system 1 according to this embodiment provides an information processing device, a dielectric loss tangent prediction method, and a program that can predict the Df of a target compound with less computational cost and higher accuracy.

[0067] [Other Embodiments] The server device 10 according to this embodiment may be configured to perform the above processing in cooperation with multiple computers. For example, the server device 10 according to this embodiment may be configured to perform the above processing in cooperation with a client terminal 12.

[0068] The Df predicted in this embodiment can be used to select a combustion agent to be added to hydrocarbons for the purpose of suppressing combustion. Halogen-based combustion agents have sometimes been avoided due to concerns about their impact on the environment. On the other hand, phosphorus-based flame retardants are attracting attention as halogen-free flame retardants. The Df predicted in this embodiment can be used to select a phosphorus-based combustion agent to be added to hydrocarbons for the purpose of suppressing combustion.

[0069] Furthermore, the predicted Df in this embodiment can be used to control manufacturing equipment and other devices in which a flame retardant to be added to hydrocarbons is selected by the operator specifying Df.

[0070] Although the present invention has been described above based on examples, the present invention is not limited to the above examples, and various modifications are possible within the scope of the claims. This application claims priority to Basic Application No. 2024-159875 filed with the Japan Patent Office on September 17, 2024, the entire contents of which are incorporated herein by reference.

[0071] 1 Information Processing System 10 Server Device 12 Client Terminals 18 Communication Network 24 Molecular Structure Calculation Unit 26 Calculation Unit 28 Prediction Unit 30 Regression Model Creation Unit 32 Output Unit

Claims

1. An information processing apparatus comprising: a calculation means for calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction means for predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and molecular volume and the dielectric loss tangent; and an output means for outputting the predicted dielectric loss tangent.

2. The information processing apparatus according to claim 1, further comprising: a regression model creation means for creating the regression model using the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state as explanatory variables, and the dielectric loss tangent of the target compound as the objective variable.

3. The information processing apparatus according to claim 2, further comprising: molecular structure calculation means for calculating the molecular structure in which the predicted compound or the learned compound reaches the stable state.

4. The information processing apparatus according to claim 3, wherein the molecular structure calculation means calculates the molecular structure that is in the stable state using quantum chemical calculations or classical force field calculations.

5. The regression model described above is: Here, Df is the measured dielectric loss tangent, S is the molecular surface area, V is the molecular volume, and C ｉ The information processing apparatus according to any one of claims 1 to 4, wherein (i = 0 to 5) is the regression coefficient.

6. The information processing apparatus according to any one of claims 1 to 5, wherein the predictable compound is a flame retardant added to a substrate resin.

7. A dielectric loss tangent prediction method comprising: a calculation procedure for calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction procedure for predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and molecular volume and the dielectric loss tangent; and an output procedure for outputting the predicted dielectric loss tangent.

8. A program that causes an information processing device to execute: a calculation step of calculating the molecular surface area and molecular volume of a target compound whose molecular structure is in an energetically stable state; a prediction step of predicting the dielectric loss tangent of the target compound using a regression model that represents the correlation between the molecular surface area and molecular volume and the dielectric loss tangent; and an output step of outputting the predicted dielectric loss tangent.

Images