Information processing device, information processing method, and information processing program

Abstract

To calculate the primary structure of polymers with low computational cost and high accuracy. [Solution] The information processing device according to the embodiment comprises at least one memory and at least one processor. The at least one processor acquires information on a plurality of reactions relating to the formation of a polymer, generates at least one initial conformation for a plurality of candidate reactions corresponding to the plurality of reactions based on the information on the plurality of reactions, calculates at least one reaction parameter for each of the plurality of reactions based on the initial conformation and a machine learning potential, and calculates the primary structure of the polymer using the reaction parameter.

Description

[Technical Field]

[0001] Embodiments of this disclosure relate to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] For polymers with structures formed by linking monomers, the primary structure, including the number and distribution of monomers, the order of monomers, optical isomers formed by monomers, and the composition of monomers, is known to significantly influence the properties of the polymer. Therefore, a method for predicting the primary structure is desired for more efficient research before polymerization, the process of synthesizing polymers from monomers.

[0003] In many polymerization reactions, multiple elementary reactions proceed simultaneously, and therefore the primary structure depends on the rates of these elementary reactions. For this reason, predicting the primary structure in advance involves obtaining reaction parameters for each elementary reaction, such as activation energy and frequency factor, and then solving the reaction rate equation using these parameters. While activation energy and frequency factor for common monomers are listed in polymer handbooks, these are not always sufficient for practical purposes.

[0004] On the other hand, experimentally obtaining reaction parameters requires specialized methods such as pulsed laser polymerization (PLP) and size exclusion chromatography (SEC) (PLP-SEC) or radical capture, resulting in extremely high measurement costs. Furthermore, while reaction parameters are also estimated through simulation, these simulations primarily utilize electronic state simulations such as density functional theory (DFT), leading to high computational costs and limiting the number of elementary reactions that can be calculated. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Mikiya Fujii, "Development of Closed Loop Materials Using Digital Technology," 17th Materials Science Workshop (Shared Excerpt), 2024 / 2 / 6, URL: https: / / www.hpci-office.jp / documents / workshop / ws_material_240206_fujii.pdf [Non-Patent Document 2] Takeshi Hasegawa, "Synthesis of High Molecular Weight Acrylic Polymers Using Free Radical Solution Polymerization - Relationship between Polymerization Conditions and Molecular Weight," Toagosei Group Research Annual Report, TREND 2014 No. 17, pp. 3-10, URL: https: / / www2.toagosei.co.jp / develop / trend / Reversible Chain Transfer Catalyzed PolymerizationNo17 / no17_1.pdf [Non-Patent Document 3] P. Vana, and A. Goto, “Kinetic simulations of reversible chain transfer catalyzed polymerization (RTCP): Guidelines to optimum molecular weight control,” Macromolecular Theory Simulations 19(1), 24-35 (2010) [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] The problem this disclosure aims to solve is to calculate the primary structure of a polymer at low computational cost without conducting experiments. [Means for solving the problem]

[0007] The information processing device according to the embodiment comprises at least one memory and at least one processor. The at least one processor acquires information on a plurality of reactions relating to the formation of a polymer, generates at least one initial conformation for a plurality of candidate reactions corresponding to the plurality of reactions based on the information on the plurality of reactions, calculates at least one reaction parameter for each of the plurality of reactions based on the initial conformation and a machine learning potential, and calculates the primary structure of the polymer using the reaction parameter. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram showing an example of the hardware configuration of an information processing device according to the embodiment. [Figure 2] Figure 2 shows an example of a functional block in a processor according to this embodiment. [Figure 3] Figure 3 shows an example of a reaction list according to the embodiment. [Figure 4] Figure 4 shows an example of the structural formulas of the types of molecules involved in the reaction, relating to an embodiment. [Figure 5] Figure 5 shows an example of a candidate reaction product produced by the decomposition of the initiator, the initiation reaction, the growth reaction, the chain transfer reaction, and the termination reaction according to an embodiment. [Figure 6] Figure 6 shows an example of the generation of a reaction candidate product related to the row with ID POLY0010 in the reaction list, according to an embodiment. [Figure 7] Figure 7 shows an example of the generation of multiple initial conformations in the initial structure after the reaction, relating to an embodiment. [Figure 8] Figure 8 shows an example of a bond length scan and transition state search in an embodiment. [Figure 9] Figure 9 shows an example of the results of processing by the reaction parameter calculation unit according to an embodiment. [Figure 10]FIG. 10 is a diagram showing an example of the activation energy generated for radicals, the probability of becoming the same optical isomer, the ratio of the same optical isomers being linked (isotactic), and the ratio of the same optical isomers being linked (syndiotactic) according to an embodiment. [Figure 11] FIG. 11 is a diagram showing an example of four structures (I-A-B*, I-A-A*, I-B-A*, I-B-B*) generated by respective activation energies (Ea1, Ea2, Ea3, Ea4) for radicals I-A* and I-B*, and four generation probabilities corresponding to the four structures according to an embodiment. [Figure 12] FIG. 12 is a flowchart showing an example of a procedure for activation energy calculation processing according to an embodiment. [Figure 13] FIG. 13 is a flowchart showing an example of a procedure for activation energy calculation processing according to an embodiment. [Figure 14] FIG. 14 is a flowchart showing an example of a procedure for a process of estimating stereoregularity as a primary structure of a polymer (hereinafter referred to as stereoregularity estimation processing) according to an embodiment. [Figure 15] FIG. 15 is a flowchart showing an example of a procedure for free energy calculation processing according to a first application example of an embodiment. [Figure 16] FIG. 16 is a diagram showing an example of an extended ensemble molecular dynamics simulation by metadynamics according to a first application example of an embodiment. [Figure 17] FIG. 17 is a flowchart showing an example of a procedure for a process of estimating monomer conversion rate, polymer molecular weight, and polymer molecular weight distribution as a primary structure of a polymer according to a first application example of an embodiment.

Mode for Carrying Out the Invention

[0009] Hereinafter, embodiments will be described in detail with reference to the drawings.

[0010] (Embodiment) Figure 1 is a block diagram showing an example of the hardware configuration of the information processing device 1 according to this embodiment. As shown in Figure 1, the information processing device 1 may be connected to an external device 9A via a communication network 5. The information processing device 1 may also include an external device 9B connected via a device interface 39. The information processing device 1 receives a list of multiple reactions related to the formation of polymers composed of multiple atoms, which are input by the user. The list of reactions consists of notation according to the type of molecule, such as initiator, initiator radical, monomer, polymer, and chain transfer agent, and various types of elementary reactions, such as initiator decomposition, initiation, propagation, chain transfer, and termination reactions.

[0011] The notation used is, for example, the SMILES (Simplified Molecular Input Line Entry System) notation entered by the user. SMILES notation represents information about a specific molecule (information about atoms and how they are connected) according to certain rules. For example, SMILES notation for methane represents granular information such as one carbon (C) being connected to four hydrogen (H) atoms. Note that the notation is not limited to SMILES notation; other known notations may be used as long as the type of molecule and substance can be uniquely identified. Other notations include, for example, SMARTS (SMiles ARbitrary Target Specification) notation. To make the explanation more concrete, the types of molecules entered by the user via the input device described later will be explained assuming they are written in SMILES notation.

[0012] The information processing device 1 includes a computer 30 and an external device 9B connected to the computer 30 via a device interface 39. The computer 30, as an example, includes a processor 31, a main memory 33, an auxiliary memory 35, a network interface 37, and a device interface 39. The information processing device 1 may be implemented as a computer 30 in which the processor 31, the main memory 33, the auxiliary memory 35, the network interface 37, and the device interface 39 are connected via a bus 41.

[0013] The computer 30 shown in Figure 1 has one of each component, but it may have multiple identical components. Also, although Figure 1 shows one computer 30, the software may be installed on multiple computers, and each of these multiple computers may execute the same or different parts of the software's processing. In this case, it may be a distributed computing configuration in which each computer communicates via a network interface 37 or the like to execute processing. In other words, the information processing device 1 in this embodiment may be configured as a system that realizes the various functions described later by having one or more computers execute instructions stored in one or more storage devices. Furthermore, information transmitted from a terminal may be processed by one or more computers located on the cloud, and the processing results may be transmitted to a terminal such as a display device (display unit) corresponding to an external device 9B.

[0014] The various calculations performed by the information processing device 1 in this embodiment may be executed in parallel using one or more processors, or using multiple computers connected via a network. Alternatively, the various calculations may be distributed to multiple processing cores within a processor and executed in parallel. Furthermore, some or all of the processing and means of this disclosure may be executed by at least one of a processor and a storage device located on a cloud that can communicate with computer 30 via a network. Thus, the various methods described later in this embodiment may take the form of parallel computing using one or more computers.

[0015] The processor 31 may be an electronic circuit (processing circuit, processing circuitry, CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), or ASIC (Application Specific Integrated Circuit), etc.) including the control unit and arithmetic unit of the computer 30. Alternatively, the processor 31 may be a semiconductor device including a dedicated processing circuit. The processor 31 is not limited to an electronic circuit using electronic logic elements, but may also be realized by an optical circuit using optical logic elements. Furthermore, the processor 31 may include arithmetic functions based on quantum computing.

[0016] The processor 31 performs calculations based on data and software (programs) input from various devices within the computer 30, and can output calculation results and control signals to these devices. The processor 31 may also control the various components of the computer 30 by executing the computer 30's OS (Operating System) or applications.

[0017] The information processing device 1 in this embodiment may be implemented by one or more processors 31. Here, the processor 31 may refer to one or more electronic circuits arranged on one chip, or to one or more electronic circuits arranged on two or more chips or two or more devices. When multiple electronic circuits are used, each electronic circuit may communicate by wire or wireless.

[0018] The main memory 33 is a storage device that stores instructions executed by the processor 31 and various data, and the information stored in the main memory 33 is read by the processor 31. The auxiliary storage device 35 is a storage device other than the main memory 33. These storage devices refer to any electronic component capable of storing electronic information, and may be semiconductor memory. The semiconductor memory may be either volatile memory or non-volatile memory. The storage device for storing various data used in the information processing device 1 according to this embodiment may be implemented by the main memory 33 or the auxiliary storage device 35, or by the built-in memory of the processor 31. For example, the storage unit in this embodiment may be implemented by the main memory 33 or the auxiliary storage device 35.

[0019] Multiple processors may be connected to one memory device, or a single processor 31 may be connected to it. Multiple memory devices may be connected to one processor. If the information processing device 1 in this embodiment consists of at least one memory device and multiple processors connected to this at least one memory device, it may include a configuration in which at least one of the multiple processors is connected to at least one memory device. This configuration may also be realized by memory devices and processors 31 included in multiple computers. Furthermore, it may include a configuration in which the memory device is integrated with the processor 31 (for example, a cache memory including an L1 cache and an L2 cache).

[0020] The network interface 37 is an interface for connecting to the communication network 5 wirelessly or via a wired connection. The network interface 37 can be any appropriate interface, such as one conforming to existing communication standards. Information may be exchanged between the computer 30 and an external device 9A connected via the communication network 5 through the network interface 37. The communication network 5 may be a WAN (Wide Area Network), LAN (Local Area Network), PAN (Personal Area Network), or a combination thereof, as long as information is exchanged between the computer 30 and the external device 9A. An example of a WAN is the Internet; an example of a LAN is IEEE 802.11 or Ethernet®; and an example of a PAN is Bluetooth® or NFC (Near Field Communication).

[0021] The device interface 39 is an interface such as USB (Universal Serial Bus) that directly connects to output devices such as display devices, input devices, and external devices 9B. The output device may include a speaker or other device that outputs audio.

[0022] External device 9A is a device connected to computer 30 via a network. External device 9B is a device directly connected to computer 30.

[0023] External device 9A or external device 9B may, for example, be an input device (input unit). The input device may be, for example, a camera, microphone, motion capture device, various sensors, keyboard, mouse, or touch panel, and will provide the acquired information to the computer 30. Alternatively, external device 9A or external device 9B may be a personal computer, tablet terminal, or smartphone, or other device equipped with an input unit, memory, and processor.

[0024] Furthermore, external device 9A or external device 9B may, for example, be an output device (output unit). The output device may be a display device (display unit) such as an LCD (Liquid Crystal Display), CRT (Cathode Ray Tube), PDP (Plasma Display Panel), or organic EL (Electro Luminescence) panel, or it may be a speaker that outputs sound, etc. Also, external device 9A or external device 9B may be a device equipped with an output device, memory, and processor, such as a personal computer, tablet terminal, or smartphone.

[0025] Furthermore, external device 9A or external device 9B may be a storage device (memory). For example, external device 9A may be network storage, and external device 9B may be storage such as an HDD.

[0026] Furthermore, the external device 9A or external device 9B may be a device that has some of the functions of the components of the information processing device 1 in this embodiment. In other words, the computer 30 may transmit or receive some or all of the processing results of the external device 9A or external device 9B.

[0027] Figure 2 shows an example of a functional block implemented by one or more processors 31. The processor 31 includes, for example, an acquisition unit 311, a structure generation unit 313, a conformation generation unit 315, a reaction parameter calculation unit 317, and a primary structure calculation unit 319 as functions implemented by the processor 31. The functions implemented by the acquisition unit 311, the structure generation unit 313, the conformation generation unit 315, the reaction parameter calculation unit 317, and the primary structure calculation unit 319 are each stored as programs in, for example, the main memory 33 or the auxiliary memory 35. The processor 31 can implement the functions related to the acquisition unit 311, the structure generation unit 313, the conformation generation unit 315, the reaction parameter calculation unit 317, and the primary structure calculation unit 319 by reading and executing each program stored in the main memory 33 or the auxiliary memory 35.

[0028] The acquisition unit 311 acquires information on multiple reactions related to polymer formation. For example, as an example of the information on multiple reactions, the acquisition unit 311 acquires a list of multiple reactions related to the formation of the polymer to be analyzed (hereinafter referred to as the reaction list). The multiple reactions correspond to multiple elementary reactions related to polymer formation. The multiple elementary reactions correspond to at least one of the following: initiator decomposition, initiation reaction, growth reaction, chain transfer reaction, and termination reaction. The acquisition unit 311 acquires the reaction list entered by the user via the input device from the input device. Note that the acquisition of the reaction list by the acquisition unit 311 is not limited to acquisition from the input device, but may also be acquired from a terminal device corresponding to external device 9A via the communication network 5, or from an external memory corresponding to external device 9B via the device interface 39.

[0029] Figure 3 shows an example of a reaction list RL. As shown in Figure 3, the reaction list RL is entered in, for example, CSV (Comma Separated Values) format. However, since the input of the reaction list RL is performed by the user, it is not limited to CSV format and may be entered in other formats. Multiple elementary reactions in the reaction list RL shown in Figure 3 are distinguished, for example, by an identifier (ID: Identification) that identifies the multiple elementary reactions. That is, the ID column in the reaction list RL shown in Figure 3 shows a string for identifying multiple elementary reactions. The reaction list RL shown in Figure 3 shows, for example, a list of multiple reactions (elementary reactions) that can be considered when producing each of multiple polymers. Note that the reaction list may also be a list that contains only the information of the reactions that the user wishes to calculate.

[0030] In the reaction list RL shown in Figure 3, the Radical_SMILES column shows the radical to be reacted in the elementary reaction (hereinafter referred to as the pre-reaction radical) in SMILES format. The Monomer_SMILES column in the reaction list RL shown in Figure 3 shows the monomer that reacts with the pre-reaction radical in the elementary reaction, in SMILES format. The Reaction_Type column in the reaction list RL shown in Figure 3 indicates the type of elementary reaction. Initiation indicates the initiation reaction, and Transfer indicates a chain transfer reaction. The Transfer_agent_SMILES column in the reaction list RL shown in Figure 3 shows the chain transfer agent that reacts with the pre-reaction radical in the chain transfer reaction (Transfer).

[0031] For example, in the reaction list RL shown in Figure 3, the row with ID POLY00001 indicates that the pre-reaction radical is N#C[C](C)(C), the monomer that reacts with the pre-reaction radical is CC(=C)C(=O)OC, and the elementary reaction is the initiation reaction. Also, for example, in the reaction list RL shown in Figure 3, the row with ID POLY00003 indicates that the pre-reaction radical is N#C[C](C)(C), the chain transfer agent is O=C(O)CCS, and the elementary reaction is a chain transfer reaction.

[0032] Figure 4 shows an example of the structural formulas of the types of molecules involved in the reaction. In Figure 4, AIBN (azobisisobutyronitrile) is shown as the initiator. However, the initiator is not limited to AIBN, and may be an azo radical initiator (azo compound) and / or other compounds. The monomers and polymers shown in Figure 4 are examples, and the monomers and polymers entered into the reaction list RL are not limited to these.

[0033] The structure generation unit 313 generates structures for multiple reaction candidates based on information about multiple reactions. For example, the structure generation unit 313 generates structures for reaction candidates (reaction candidates) for each of the multiple reactions in the polymer formation process based on the reaction list RL. The structures of the reaction candidates are represented, for example, as two-dimensional data. For example, the structure generation unit 313 generates reaction candidates using the types of elementary reactions and the target molecules in the reaction list RL according to pre-set rules (rule-based). These rules are defined in a look-up table (LUT) that shows reaction candidates for the types of elementary reactions and the target molecules, and are stored in the main memory 33 or auxiliary memory 35. Note that the rule-based approach is not limited to the above look-up table (LUT), and may be implemented by various programs (calculation programs, calculation processes, neural networks) that calculate reaction candidates using the types of elementary reactions and the target molecules as input. Furthermore, when predicting reaction candidates using a neural network, a neural network corresponding to a diffusion model, for example, may be used. The input to the diffusion model is not limited to the SMILES format; it may also be the name of the compound. Based on the above, the structure generation unit 313 enumerates the reaction candidates using the elements in the reaction list RL.

[0034] Figure 5 shows an example of a candidate reaction product generated by the decomposition of the initiator, the initiation reaction, the growth reaction, the chain transfer reaction, and the termination reaction. In Figure 5, it is shown that as initiator decomposition, initiator radicals are generated by uniform cleavage in response to heat treatment of the initiator AIBN by temperature Δ and / or light irradiation by photon flux density hν. For example, the structure generation section 313 lists structures generated by the decomposition of nitrogen molecules N=N or COOC with respect to initiator decomposition.

[0035] Furthermore, Figure 5 shows the reaction between an initiator radical and a monomer as the initiation reaction. In the initiation reaction, α and β indicate the positions of carbon atoms in the monomer to which the initiator radical can be added. Accordingly, the structure generation unit 313 lists the structures generated by the bonding of the initiator radical to the carbon atom at position α or position β in the monomer, that is, the structures of candidate reaction products generated by the addition of the carbon atom at position α or β to the initiator radical.

[0036] Furthermore, Figure 5 shows a growth reaction in which a grown radical, formed by adding at least one monomer to an initiator, reacts with a monomer. As a result, the structure generation unit 313 lists the structures generated when the carbon at the α or β position of the monomer is added to the grown radical from the front side of the page (represented as a wedge shape in the structural formula) or the back side of the page (represented as a dashed line in the structural formula), i.e., the structures of reaction candidates corresponding to the length of the grown radical, the α and β positions, and the optical isomers.

[0037] Furthermore, Figure 5 shows a chain transfer reaction in which a radical abstracts a hydrogen atom from a polymer. The structure generation unit 313 then lists the structures of the reaction candidates after the radical abstracts a hydrogen atom. As shown in Figure 5, the reaction candidates are generated as two-dimensional structures (e.g., structural formulas).

[0038] Furthermore, Figure 5 shows examples of two types of termination reactions. The recombination termination shown in Figure 5 is an example of a termination reaction due to the recombination of radicals between polymers. The disproportionation termination shown in Figure 5 is a termination reaction due to the formation of a double bond (disproportionation) by the abstraction of hydrogen from the radical. Based on this, the structure generation unit 313 lists the structures of reaction candidates resulting from either recombination termination or disproportionation termination.

[0039] Figure 6 shows an example of the generation SG of a candidate reaction product for the row with ID POLY00010 in the reaction list RL. As shown in Figure 6, the structure of the grown radical in the row with ID POLY00010 is represented, for example, α(R)-α(R)-α(R), β(R)-α(R)-α(R), etc. Here, R (rectus) in the structure of the grown radical indicates dextrorotation in the optical isomer. Levorotation in the optical isomer is denoted by S (sinister). Note that dextrorotation may be denoted as D (Dextrorotary(+)) and levorotation as L (Levorotary(-)).

[0040] The addition sites shown in Figure 6 correspond, for example, to the carbon atoms at the α or β positions in the monomer shown in Figure 5. Furthermore, the initial structure of the reaction candidate after the reaction shows the structure of the reaction candidate generated by the structure generation unit 313. The initial structure of the reaction candidate is generated regardless of, for example, the direction (angle) of rotation with the direction of various bonds as the axis of rotation, or three-dimensional twisting. In other words, the initial structure of the reaction candidate does not have three-dimensional coordinates. For example, in Figure 6, multiple reaction candidate structures are generated depending on the structure of multiple growth radicals before the reaction, the addition site (α or β), and the optical isomer (R or S). As shown in Figure 6, for example, if the structure of the growth radical is α(R)-α(R)-α(R) and β(R)-α(R)-α(R), the addition site is β, and the optical isomer is S, the structure generation unit 313 generates the initial structure ARS after the reaction.

[0041] The conformation generation unit 315 generates at least one initial conformation for a candidate reaction based on the structure of the generated candidate reaction. A conformation corresponds to a three-dimensional coordinate system that reproduces the angles, twists, etc., of the bonding between multiple elements in the candidate reaction. The conformation generation unit 315 generates (determines) at least one initial conformation, for example, using a classical force field based on the structure of the candidate reaction. The initial conformation corresponds to three-dimensional data that shows the three-dimensional coordinates (a group of three-dimensional coordinates) of the multiple atoms constituting the candidate reaction. Using a classical force field is equivalent to imposing physically reasonable constraints on the generation of the initial conformation, such as preventing overlap between three-dimensional atoms. Furthermore, the generation of the initial conformation is not limited to using classical force fields; it may also be generated using a pre-trained neural network (hereinafter referred to as Neural Network Potential: NNP) that predicts the total energy of the atomic state and the forces acting on each individual atom. For example, the Experimental Torsion angle preferences and basic Knowledge Distance Geometry (ETKDG) (see URL: https: / / pubs.acs.org / doi / abs / 10.1021 / acs.jcim.5b00654) may be used as the pre-trained neural network that predicts the total energy of the atomic state and the forces acting on each individual atom.

[0042] Figure 7 shows an example of generating multiple initial conformations for the initial structure ARS after a reaction. As shown in Figure 7, the conformation generation unit 315 generates multiple initial conformations based on the initial structure ARS after a reaction, with monomer angles, twists, etc., differing within a predetermined range. As shown in Figure 7, the initial conformations correspond to a three-dimensional representation of the reaction candidate. Note that the generation of initial conformations is not limited to the above description. For example, a machine learning model may be used that generates the initial conformation after a reaction when pre-reaction SMILES (information about the reaction) is input. The initial conformation after a reaction is, for example, an initial conformation that corresponds to the structure of the reaction candidate predicted from the pre-reaction SMILES. The machine learning model may be the diffusion model (neural network) described above. Also, if the information about multiple reactions includes the names of compounds (i.e., pre-reaction SMILES is replaced with names), the machine learning model may be configured to output a set of three-dimensional coordinates (initial conformations) from the names of the compounds. In these cases, the structure generation unit 313 is not required. According to the structure generation unit 313 and the conformation generation unit 315, at least one initial conformation is generated for a plurality of candidate reaction corresponding to a plurality of reactions, based on information about a plurality of reactions.

[0043] For example, in the reaction list RL shown in Figure 6, in the generation SG of the reaction candidate in the row with ID POLY00010, the structure that can be generated by the growth reaction (Propgation) by a combination of two SMILES is expressed as, for example, the structure of the growth radical × the addition site (α, β) of the next monomer × the next chirality (R, S). Here, the structure of the growth radical is determined by the allowed degree of polymerization, the addition site, and the chirality. For example, if the α carbon is the addition site and the β carbon is the addition site, and both R and S are considered as chiralities, the number of growth radicals will be (2 × 2)^(degree of polymerization). If the direction of rotation (angle) with respect to the direction of various bonds as the axis of rotation, and the three-dimensional twist are expressed in 10 patterns for the structure that can be generated, then the number of initial conformations will be the structure of the growth radical × the addition site of the next monomer × the next chirality × 10 conformations.

[0044] The reaction parameter calculation unit 317 calculates at least one reaction parameter for each of the multiple reactions based on the initial conformation and the machine learning potential. For example, the reaction parameter calculation unit 317 calculates the reaction parameters for each of the multiple reactions using a trained model that predicts the total energy of the atomic states of the candidate reaction material and the forces acting on each individual atom, along with the generated initial conformation. Specifically, the reaction parameter calculation unit 317 calculates the reaction parameters for each of the multiple reactions using the structural optimization of the material in the polymer formation process using the trained model and at least one initial conformation. The trained model is, for example, an NNP. The reaction parameters for each of the multiple reactions related to the formation of the polymer under analysis include at least one of the following: activation energy, activation free energy (hereinafter referred to as free energy), frequency factor, and reaction rate constant.

[0045] For example, when calculating the activation energy as a reaction parameter, the reaction parameter calculation unit 317 generates multiple pre-reaction conformations using the initial conformation based on structural optimization using a machine learning potential, and calculates the activation energy by performing a transition state search using the machine learning potential with these multiple pre-reaction conformations. The machine learning potential is, for example, a neural network potential. The neural network potential is a neural network that outputs energy when an atomic structure is input. The machine learning potential may further have a configuration that calculates force using the position derivative of energy. The machine learning potential may also have a configuration that outputs force when an atomic structure is input. Furthermore, the machine learning potential may be implemented, for example, by a trained model that predicts the total energy of the atomic states of a candidate reaction product and the force experienced by each individual atom in the candidate product.

[0046] For example, the conformation generation unit 315 generates multiple initial conformations based on the structure of the reaction candidate. Next, the reaction parameter calculation unit 317 calculates the activation energy for each of the multiple initial conformations and a trained model. Specifically, for each of the multiple initial conformations, the reaction parameter calculation unit 317 calculates multiple pre-reaction conformations (hereinafter referred to as pre-reaction conformations) by scanning bond lengths (bond distances) accompanied by structural optimization using the trained model. The bond length scan may also be called conformational exploration because it searches for pre-reaction conformations by extending the bond lengths. For each of the multiple initial conformations, the reaction parameter calculation unit 317 searches for the transition state from before the reaction to after the reaction using the calculated pre-reaction conformations, accompanied by structural optimization using the trained model. As a result, the reaction parameter calculation unit 317 calculates multiple activation energies corresponding to each of the multiple initial conformations. The calculation of free energy will be explained in the first application example described later.

[0047] Figure 8 shows an example of bond length scanning and transition state exploration. As shown in Figure 8, the reaction parameter calculation unit 317 extends the bond distance by a predetermined distance in the initial conformational ICM. For example, the reaction parameter calculation unit 317 generates multiple pre-reaction conformations (pre-reaction conformations) by performing structural optimization while changing the bond length. Bond length scanning (bond distance scanning) is the process of performing structural optimization while changing the bond length between atoms. In bond distance scanning, the reaction parameter calculation unit 317 scans (explores) pre-reaction conformations from the post-bonded state to the pre-bonded state. At this time, the determination of which bonds between atoms disappear in the conformation is based on a pre-set lookup table. Therefore, the reaction parameter calculation unit 317 automatically determines the group of atoms to move in bond distance scanning according to the lookup table. The direction in which atoms are moved in bond distance scanning corresponds to a vector (unit vector) along the bond axis to be eliminated. The final distance in a link distance scan is the distance at which the link can be considered broken, for example, 1.5 times the van der Waals distance.

[0048] For example, if the bond distance is 1.5 Å, the predetermined distance is, for example, 0.1 Å. That is, the predetermined distance is set in advance, such as 1 / 15 of the bond distance, and stored in the main memory 33 or auxiliary memory 35. The reaction parameter calculation unit 317 inputs the initial conformation in which the bond distance has been extended by the predetermined distance (hereinafter referred to as the bond extension conformation) into the trained model and performs structural optimization of the bond extension conformation. As a result, the reaction parameter calculation unit 317 generates a first conformation C1 that is equilibrated with respect to energy and force.

[0049] Next, the reaction parameter calculation unit 317 extends the bond distance in the first conformation C1 by a predetermined distance. The reaction parameter calculation unit 317 inputs the first conformation C1, with the bond distance further extended by the predetermined distance, into the trained model and performs structural optimization of the bond-extended conformation. The reaction parameter calculation unit 317 may also perform structural optimization by adding forces that extend the bond length. As a method for structural optimization, for example, known optimization algorithms such as the L-BFGS method (Limited-memory BFGS (Broyden-Fletcher-Goldfarb-Shanno) method) are used to perform structural optimization. Through these steps, the reaction parameter calculation unit 317 generates a second conformation C2 that is equilibrated with respect to energy and force. Scanning the bond length corresponds to repeating the above process.

[0050] In other words, a bond extension conformation with an extended bond distance is input to a trained model, and a new bond extension conformation is generated by optimizing the structure of the bond extension conformation. The bond length scan is repeated until the bond distance extends to a set value. The set value is predetermined, such as several times the bond distance, and is stored in the main memory 33 or auxiliary memory 35.

[0051] Figure 8 shows eight bond elongation conformations (CECs), but is not limited to these, and many more bond elongation conformations may be generated. In the example shown in Figure 8, nine conformations (initial conformation ICM and eight bond elongation conformations) are shown. The reaction parameter calculation unit 317 uses the bond elongation conformations CEC and initial conformation ICM generated by scanning the bond lengths to search for the transition state. In the example shown in Figure 8, the bond elongation conformations CEC and initial conformation ICM correspond to the initial conditions for searching for the transition state.

[0052] The algorithm for searching for transition states can be any known method, such as the Nudged Elastic Band method or the String method, so a detailed explanation is omitted. In searching for transition states, a trained model is used to evaluate the forces and / or energies in the modified conformation.

[0053] For example, as shown in Figure 8, the transition state search in the reaction parameter calculation unit 317 calculates the conformational RCT of the reactant before the reaction, the conformation of the reactant in the transition state TS, and the conformational PR of the reaction product. At this time, the transition state search also calculates the conformation of the reactant on the reaction pathway between the conformational RCT of the reactant before the reaction and the transition state TS, and the conformation on the reaction pathway between the transition state TS and the reaction product PR (hereinafter referred to as the intermediate conformation). The reaction pathway RP in Figure 8 corresponds to the pathway in the Free Energy Surface (FES) for degrees of freedom that is most likely to cause a reaction. Degrees of freedom are, for example, the interatomic distance between two atoms to be bonded (Reaction Coordinate, or Collective Variable (CV)).

[0054] For example, the reaction parameter calculation unit 317 inputs the conformational RCT of the reactants before the reaction, the conformation of the reactants in the transition state TS, the conformational PR of the reaction products, and the intermediate conformation into a trained model, and performs structural optimization of these conformations. Next, the reaction parameter calculation unit 317 calculates the bond distances related to the reaction and the distances between atoms other than the bond distances for each of the multiple conformations of the reactants whose structures have been optimized along the reaction pathway RP.

[0055] The reaction parameter calculation unit 317 calculates the activation energy, which is the energy difference between the transition state TS and the conformation of the reactants before the reaction, by searching for the transition state TS. For example, the reaction parameter calculation unit 317 calculates the activation energy based on the conformation of the transition state TS after structural optimization and the conformation of the reactants before the reaction after structural optimization.

[0056] Next, the reaction parameter calculation unit 317 determines whether or not each of the multiple initial conformations is unnecessary as an activation energy. For example, the reaction parameter calculation unit 317 compares the bond distance related to the reaction and the distances between atoms other than the bond distance along the reaction pathway RP with predetermined conditions. The predetermined conditions correspond to, for example, that the bond distance related to the reaction has changed and that the distances between atoms other than the bond distance have not changed by more than a predetermined value. For example, a predetermined condition is that if an atom is within 60% of the van der Waals radius, it is determined to be bonded (except between hydrogen atoms). Based on these, the reaction parameter calculation unit 317 determines for each of the multiple initial conformations whether or not the bond distance related to the reaction has changed and whether or not the distances between atoms other than the bond distance have changed by more than a predetermined value.

[0057] If the bond distance related to the reaction has not changed, or if the distance between atoms other than the bond distance has changed by more than a predetermined value, the reaction parameter calculation unit 317 deletes various conformations, such as the initial conformation to be judged, as unnecessary reactions. In other words, the reaction parameter calculation unit 317 determines for each of the multiple initial conformations whether predetermined conditions are met in the reaction process. The predetermined conditions are set in advance and stored in the main memory 33 or auxiliary memory 35.

[0058] Figure 9 shows an example of the results of processing by the reaction parameter calculation unit 317. As shown in Figure 9, the reaction in initial conformation A is removed as an unnecessary reaction. Note that the activation energy may be calculated for multiple initial conformations after determination using predetermined conditions.

[0059] The primary structure calculation unit 319 calculates the primary structure of the polymer using reaction parameters. The primary structure is, for example, at least one of the following: stereoregularity, monomer chain distribution, head-to-tail bonding, and molecular weight distribution. Stereoregularity, monomer chain distribution, head-to-tail bonding, and molecular weight distribution may be referred to as monomer sequence. That is, the primary structure calculation unit 319 calculates (simulates) the monomer sequence using activation energy or free energy. The primary structure calculation unit 319 may also be referred to as the monomer sequence calculation unit.

[0060] Stereoregularity describes how the repeating structural units (e.g., monomers) that make up a polymer are linked together in a sterically regular and continuous manner. Stereoregularity is classified into categories such as isotactic and syndiotactic.

[0061] The primary structure calculation unit 319 calculates the probability of a chain of identical optical isomers occurring using a Boltzmann distribution based on the activation energy for bonding identical optical isomers and the activation energy for bonding different optical isomers. Stereoregularities such as the probability of a chain of identical optical isomers occurring are reflected in the properties of the polymer.

[0062] Figure 10 shows an example of the activation energy for the structure generated in response to a radical, the probability pm of forming the same optical isomer, the proportion I of identical optical isomers linked together (isotactic), and the proportion S of identical optical isomers linked together (syndiotactic). In the table in Figure 10, the radical column shows, as an example, a molecule in which a dextrorotatory monomer MMA(R) and a radical MMA* are bound to initiator I, and a molecule in which a levorotatory monomer MMA(S) and a radical MMA* are bound to initiator I.

[0063] Furthermore, the table in Figure 10 shows that there are two types of structures generated from the radical I-MMA(R)-MMA*: I-MMA(R)-MMA(R)-radical MMA* and I-MMA(R)-MMA(S)-radical MMA*. Also, the table in Figure 10 shows that there are two types of structures generated from the radical I-MMA(S)-MMA*: I-MMA(S)-MMA(R)-radical MMA* and I-MMA(S)-MMA(S)-radical MMA*.

[0064] Furthermore, in the table shown in Figure 10, the activation energy Ea, which is the result of processing by the reaction parameter calculation unit 317, is shown for each of the generated structures. By performing a calculation with the partition function as the denominator using the Boltzmann factor calculated using the activation energy Ea, the probability P m The probability P is calculated. m Using this, the proportion of isotactic behavior I and the proportion of syndiotactic behavior S are calculated.

[0065] For example, suppose a polymer is produced using I as an initiator and M as a monomer, represented as IMMMMMMMMMMMMMMM. In this case, if the polymer is represented as RRRSSRRRSSRRRRR, where S represents the levorotatory nature of monomer M and R represents the dextrorotatory nature of monomer M, the proportion of isotactic polymers I will be large because the same optical isomers tend to be consecutive. On the other hand, if the polymer is represented as RSRSRSRSSRSSRS, the proportion of syndiisotactic polymers S will be large because different optical isomers tend to be consecutive.

[0066] The monomer chain distribution corresponds to the distribution of the arrangement of multiple monomers in a polymer. For the purpose of this explanation, let's assume that the multiple monomers that make up the polymer consist of two types of monomers (monomer A and monomer B). In this case, the monomer chain is generated when monomer A or monomer B is added to a radical (bonded by reaction). The probability k of the reaction between monomer A or monomer B and the radical is calculated using the following formula with activation energy Ea: k = exp(-Ea / (RT)). The probability k of the reaction may also be called the reaction probability. In the formula showing the probability k, R is the gas constant and T is the temperature during the reaction. The probability k may also be calculated using the following formula with free energy Ga: k = exp(-Ga / (RT)).

[0067] Figure 11 shows an example of four structures (IAB*, IAA*, IB-A*, IB-B*) generated for radicals IA* and IB* with different activation energies (Ea1, Ea2, Ea3, Ea4), and the corresponding four generation probabilities. As shown in Figure 11, for radical IA*, reaction product IAB* is generated with an activation energy of Ea1 and a generation probability of 0.05. As shown in Figure 11, for radical IA*, reaction product IAA* is generated with an activation energy of Ea2 and a generation probability of 0.05. Also, as shown in Figure 11, for radical IB*, reaction product IBA* is generated with an activation energy of Ea3 and a generation probability of 0.1. As shown in Figure 11, for radical IB*, reaction product IBB* is generated with an activation energy of Ea4 and a generation probability of 0.8.

[0068] In the example shown in Figure 11, it is shown that the reaction product IBB* is more likely to occur with respect to the radical IB* (k=0.8), so in the resulting polymer, monomer B will be the majority, such as IBBAABBBBBBBBBBBBBBB. Based on the above, the primary structure calculation unit 319 calculates the chain distribution of monomers using the generation probability k, which indicates the difference in reactivity between monomers and radicals. As a result, the primary structure calculation unit 319 predicts the structure of the resulting polymer. The chain distribution of monomers contributes to the determination of higher-order structures, such as crystallinity. Differences in these higher-order structures make it possible to calculate physical properties such as processability, rigidity (flexibility), transparency, and impact strength of the polymer.

[0069] A head-to-tail bond indicates, for example, that a radical is continuously bonded to the carbon at the α or β position in a monomer during a growth reaction. The alternating bonding of radicals to the α and β positions in a monomer during a growth reaction is called head-to-head or tail-to-tail. In this case, the primary structure calculation unit 319 may calculate head-to-tail, head-to-head, and tail-to-tail bonds as primary structures. The calculation of head-to-tail, head-to-head, and tail-to-tail bonds is performed by calculating the reaction probability between the carbon at the α or β position in the monomer and the radical, using the activation energy, similar to the calculation of the monomer's chain distribution.

[0070] The molecular weight distribution corresponds to the distribution of the number-average molecular weight and / or weight-average fraction in the multiple polymers produced. For example, in radical polymerization, a polymer of a certain length is produced by the simultaneous occurrence of multiple elementary reactions. Since the molecular weight distribution is basically obtained by numerically solving differential equations using free energy, the calculation of molecular weight distribution will be explained in the application examples.

[0071] The configuration of the information processing device 1 has been described above. Below, the procedure for the activation energy calculation process (hereinafter referred to as the activation energy calculation process) performed by the information processing device 1 will be explained using Figures 12 and 13.

[0072] Figures 12 and 13 are flowcharts showing an example of the activation energy calculation procedure.

[0073] (Activation energy calculation process) (Step S111) The input device receives a reaction list RL containing multiple reactions related to the formation of the polymer to be analyzed, according to the user's instructions. For example, the input device receives the reaction list RL in CSV format, as shown in Figure 3. At this time, the acquisition unit 311 acquires the reaction list RL from the input device or the like. The acquisition unit 311 stores the acquired reaction list RL in the main memory 33 or auxiliary memory 35.

[0074] (Step S112) The structure generation unit 313 generates structures of reaction candidate products for each of multiple reactions based on the reaction list RL. For example, the structure generation unit 313 generates structures of reaction candidate products from the reaction list RL using a correspondence table (LUT) or calculation program that shows the types of elementary reactions and reaction candidate products for the target molecules. The structure generation unit 313 stores the generated structures of reaction candidate products in the main memory 33 or auxiliary memory 35.

[0075] (Step S113) The conformation generation unit 315 generates multiple initial conformations for a candidate reaction based on the structure of the generated candidate reaction. For example, the conformation generation unit 315 generates multiple initial conformational ICMs as shown in Figure 7, using a trained model such as a classical force field or NNP based on the structure of the candidate reaction, i.e., the initial structure ARS after the reaction. The conformation generation unit 315 stores the generated initial conformational ICMs in the main memory 33 or the auxiliary memory 35.

[0076] (Step S114) The reaction parameter calculation unit 317 extends the bond length related to the preceding reaction in the initial conformation to a predetermined length. That is, in the initial structure ARS after the reaction, the reaction parameter calculation unit 317 extends the bond length (bond distance) between the two atoms related to the reaction to a predetermined length. The reaction parameter calculation unit 317 stores the initial conformation (bond extension conformation) having the bond distance extended to the predetermined length in the main memory 33 or auxiliary memory 35.

[0077] (Step S115) The reaction parameter calculation unit 317 inputs the bond extension conformation, in which the bond length is extended, into a trained model such as an NNP and calculates the energy in the bond extension conformation. Next, the reaction parameter calculation unit 317 uses the calculated energy to calculate the force in the bond extension conformation. The reaction parameter calculation unit 317 stores the calculated energy and force in the main memory 33 or the auxiliary memory 35.

[0078] (Step S116) The reaction parameter calculation unit 317 uses the calculated forces to perform structural optimization for the bond extension conformation. The reaction parameter calculation unit 317 may also use the calculated energy to further optimize the structure for the bond extension conformation. This updates the bond extension conformation for the reaction parameter calculation unit 317. The reaction parameter calculation unit 317 stores the updated bond extension conformation in the main memory 33 or the auxiliary memory 35.

[0079] (Step S117) The reaction parameter calculation unit 317 determines whether the bond length in the updated bond extension conformation exceeds a set value. If the bond length in the updated bond extension conformation does not exceed the set value (bond length ≤ set value, No in step S117), the process in step S118 is executed. If the bond length in the updated bond extension conformation exceeds the set value (bond length > set value, Yes in step S117), the process in step S119 is executed.

[0080] (Step S118) The reaction parameter calculation unit 317 extends the bond length in the bond extension conformation over a predetermined length. The reaction parameter calculation unit 317 stores the bond extension conformation having the bond distance further extended over the predetermined length in the main memory 33 or auxiliary memory 35. The processing in steps S114 to S118 corresponds to a bond length scan. The bond length scan generates multiple bond extension conformations CEC and initial conformations ICM generated by the bond length scan, for example, as shown in Figure 8. Specifically, multiple bond extension conformations are generated in each of the multiple initial conformations.

[0081] (Step S119) The reaction parameter calculation unit 317 searches for a transition state between the pre-reaction and post-reaction state using the initial conformation and a plurality of bond extension conformations. Through this search, the reaction parameter calculation unit 317 calculates, for example, the conformational RCT of the reactants before the reaction, the conformation of the reactants in the transition state TS, the conformational PR of the reaction product, and the intermediate conformation, as shown in Figure 8. The reaction parameter calculation unit 317 stores the conformational RCT of the reactants before the reaction, the conformation of the reactants in the transition state TS, the conformational PR of the reaction product, and the intermediate conformation in the main memory 33 or auxiliary memory 35.

[0082] (Step S120) The reaction parameter calculation unit 317 optimizes the structures of these substances by inputting the conformational RCT of the reactants before the reaction, the conformation of the reactants in the transition state TS, the conformational PR of the reaction products, and the intermediate conformation into a trained model. Specifically, the reaction parameter calculation unit 317 inputs the conformational RCT of the reactants before the reaction, the conformation of the reactants in the transition state TS, the conformational PR of the reaction products, and the intermediate conformation into the NNP and calculates the energy in each conformation. Next, the reaction parameter calculation unit 317 calculates the force in each conformation (each of the multiple conformations) using the calculated energy.

[0083] (Step S121) The reaction parameter calculation unit 317 compares each of the multiple forces corresponding to the multiple conformations with a predetermined force. At this time, the reaction parameter calculation unit 317 may also compare each of the multiple energies corresponding to the multiple conformations with a predetermined energy. The predetermined force and predetermined energy are set in advance and stored in the main memory 33 or the auxiliary memory 35.

[0084] If the calculated force exceeds a predetermined force and / or the calculated energy exceeds a predetermined energy (Yes in step S121), the process in step S122 is executed. If the calculated force is less than or equal to the predetermined force and / or the calculated energy is less than or equal to the predetermined energy (No in step S121), the process in step S124 is executed.

[0085] In other words, if the multiple forces corresponding to multiple conformations are less than or equal to a predetermined force, and / or the multiple energies corresponding to multiple conformations are less than or equal to a predetermined energy, the search for the transition state is completed. At this time, the reaction parameter calculation unit 317 stores each of the multiple conformations used in the search for the transition state in the main memory 33 or auxiliary memory 35 as a conformation related to the determination of the transition state (hereinafter referred to as the transition state determination conformation). The processing in this step corresponds to the evaluation of forces and / or energies for the changed conformations in the search for the transition state.

[0086] (Step S122) The reaction parameter calculation unit 317 uses the calculated force for each of the multiple conformations to perform structural optimization for the conformation corresponding to that force. The reaction parameter calculation unit 317 may also use the calculated energy to perform structural optimization for the conformation. In this way, the reaction parameter calculation unit 317 updates each of the multiple conformations using the force calculated for each of the multiple conformations. Note that the conformation update may be performed only for conformations corresponding to forces exceeding a predetermined force. The reaction parameter calculation unit 317 stores the updated multiple conformations in the main memory 33 or the auxiliary memory 35.

[0087] (Step S123) The reaction parameter calculation unit 317 performs a transition state search using the updated multiple conformations. The process of searching for the transition state is the same as in step S119, so the explanation is omitted.

[0088] (Step S124) The reaction parameter calculation unit 317 calculates the activation energy based on the transition state. Specifically, the reaction parameter calculation unit 317 calculates the activation energy as the energy difference between the transition state TS and the conformation of the reactants before the reaction, from among a plurality of determined transition state conformations. In other words, the reaction parameter calculation unit 317 calculates the activation energy based on the conformation of the transition state TS after structural optimization and the conformation of the reactants before the reaction after structural optimization. The reaction parameter calculation unit 317 stores the calculated activation energy in the main memory 33 or auxiliary memory 35, in association with the reaction.

[0089] (Step S125) The reaction parameter calculation unit 317 determines whether multiple transition state conformations satisfy predetermined conditions. Specifically, the reaction parameter calculation unit 317 determines whether, for each of the multiple transition state conformations on the reaction pathway PR, the bond distances related to the reaction have changed, and the distances between atoms other than the bond distances have not changed by more than a predetermined value. If the transition state conformations satisfy the predetermined conditions (Yes in step S125), the process in step S127 is executed. If the transition state conformations do not satisfy the predetermined conditions (No in step S125), the process in step S126 is executed.

[0090] (Step S126) The reaction parameter calculation unit 317 removes various conformations, such as the initial conformation to be determined, as unnecessary reactions. That is, the reaction parameter calculation unit 317 removes various conformations related to the calculated activation energy. Note that step S125 may be performed before step S124. In this case, the calculation of the activation energy for unnecessary reactions becomes unnecessary.

[0091] (Step S127) If the search for transition states has been completed for all initial conformations (Yes in step S127), the process in step S128 is executed. If the search for transition states has not been completed for all initial conformations (No in step S127), the process from step S114 onwards is executed.

[0092] (Step S128) The reaction parameter calculation unit 317 generates a list of activation energies for the target reaction. For example, the reaction parameter calculation unit 317 generates a list of activation energies for reactions targeting reaction candidate products in the reaction list RL. The reaction parameter calculation unit 317 stores the generated list of activation energies in the main memory 33 or the auxiliary memory 35.

[0093] (Step S129) If the calculation of the activation energy for all reaction candidates is not complete (No in step S129), the process from step S113 onwards is executed for the reaction candidates for which the activation energy has not been calculated. If the calculation of the activation energy for all reaction candidates is complete (Yes in step S129), the process in step S130 is executed.

[0094] (Step S130) The reaction parameter calculation unit 317 generates a list of activation energies corresponding to the reaction list. For example, the reaction parameter calculation unit 317 generates a list of activation energies corresponding to multiple reactions in the reaction list RL. The reaction parameter calculation unit 317 stores the generated list of activation energies in the main memory 33 or the auxiliary memory 35.

[0095] The above describes the activation energy calculation process performed by the information processing device 1. Below, we will describe the process of calculating (estimating) the primary structure performed by the information processing device 1. As an example, the primary structure calculation will be explained using two types of data: stereoregularity and monomer chain distribution. First, we will describe the process of estimating stereoregularity as the primary structure of the polymer (hereinafter referred to as the stereoregularity estimation process).

[0096] Figure 14 is a flowchart showing an example of the procedure for estimating three-dimensional regularity.

[0097] (Stereoscopic regularity estimation process) (Step S141) A list of monomers and their absolute conformations is input via an input device. This step corresponds to the input of a list of reactions for the polymer whose stereoregularity is to be estimated. In other words, this step corresponds to step S111 in the activation energy calculation process.

[0098] (Step S142) Based on the input list, the activation energy calculation process shown in Figures 12 and 13 is executed. This generates a list of activation energies for all reactions in the input list.

[0099] (Step S143) The primary structure calculation unit 319 uses the activation energy for each of the multiple reactions to calculate the weighting of the reaction occurrence probability (Boltzmann distribution) for the monomer repeating structure using the equation shown in Figure 10. That is, the primary structure calculation unit 319 calculates probability P as shown in Figure 10.m The primary structure calculation unit 319 calculates the reaction probability P. m The data is stored in the main memory 33 or the auxiliary memory 35.

[0100] (Step S144) The primary structure calculation unit 319 calculates the weighted (reaction probability) P m Applying this to the formula shown in Figure 10, the proportion of consecutive monomers with the same absolute configuration (isotactic proportion) I is calculated. The primary structure calculation unit 319 then calculates the weighted P m The syndiotactic ratio S is calculated by applying this to the formula shown in Figure 10. The primary structure calculation unit 319 stores the isotactic ratio I and the syndiotactic ratio S in the main memory 33 or auxiliary memory 35.

[0101] (Step S145) The primary structure calculation unit 319 estimates the stereoregularity of the polymer to be produced. This allows for the estimation of the distribution of optical isomers in the polymer. At this time, the primary structure calculation unit 319 may use the estimated stereoregularity to calculate the physical properties of the polymer to be produced.

[0102] In the process of estimating the chain distribution of monomers as the primary structure of a polymer (hereinafter referred to as the chain distribution estimation process), the number of reactions and a list of monomers are input via an input device. In the subsequent process, the process in step S142 is executed to generate a list of activation energies for all reactions related to the input list of monomers. Next, the primary structure calculation unit 319 calculates the probability of reaction occurrence k (=exp(-Ea / (RT))) which indicates the difference in reactivity between monomers and radicals, using the activation energy Ea. Then, the primary structure calculation unit 319 generates a polymer by repeating the reaction using the probability of reaction occurrence k over the number of reactions specified via the input device. The chain distribution of monomers is estimated in the generated polymer.

[0103] Based on the above, the information processing device 1 according to this embodiment acquires information (reaction list) regarding a plurality of reactions related to the formation of the polymer to be analyzed, generates a plurality of candidate reaction products corresponding to the plurality of reactions (at least one initial conformation for the reaction candidates) based on the information regarding the plurality of reactions, calculates at least one reaction parameter for each of the plurality of reactions based on the initial conformation and the machine learning potential, and calculates the primary structure of the polymer using the calculated reaction parameter. For example, the information processing device 1 according to this embodiment generates structures for a plurality of candidate reaction products based on information regarding a plurality of reactions, and generates at least one initial conformation for the candidate product based on the structure of the candidate reaction product. In the information processing device 1 according to this embodiment, the primary structure includes at least one of the monomer chain distribution, head-to-tail bonding, stereoregularity, and molecular weight distribution, and the reaction parameter includes the activation energy for each of the plurality of reactions.

[0104] Furthermore, in the information processing device 1 according to this embodiment, when calculating the activation energy as a reaction parameter, the information processing device 1 according to this embodiment generates a plurality of pre-reaction conformations using the initial conformation based on structural optimization using the machine learning potential, and calculates the activation energy by performing a transition state search using the machine learning potential with the plurality of pre-reaction conformations. In addition, the information processing device 1 according to this embodiment generates a plurality of pre-reaction conformations by performing the structural optimization while changing the bond length. Furthermore, the information processing device 1 according to this embodiment calculates the reaction probability between the growth radical or initiator and the monomer in the generation of a reaction candidate product using the activation energy. Furthermore, in the information processing device 1 according to this embodiment, the machine learning potential is a neural network potential.

[0105] Based on these findings, the information processing device 1 according to this embodiment can calculate the primary structure of a polymer by performing structural optimization in the generation process using a trained model such as NNP, and then using the trained model and the initial conformation to calculate the activation energy. As a result, the information processing device 1 according to this embodiment can calculate the activation energy without conducting experiments, at a lower computational cost compared to DFT, and with high accuracy due to structural optimization using a trained model. In addition, the information processing device 1 according to this embodiment can calculate the primary structure of any polymer at a lower cost than conducting experiments, at a lower computational cost than performing DFT, and with high accuracy due to structural optimization using NNP.

[0106] (First application example) This application involves calculating free energy (activation free energy) instead of activation energy. The calculated free energy is used to estimate the primary structure, such as stereoregularity, monomer chain distribution, and molecular weight distribution. The procedure for calculating free energy (hereinafter referred to as the free energy calculation process) will be explained below using Figure 15.

[0107] Figure 15 is a flowchart showing an example of the procedure for calculating free energy. In the flowchart shown in Figure 15, the processes in steps S151 and S152 are the same as those in steps S111 and S112, so their explanation is omitted. Therefore, the procedure for calculating free energy will be explained starting from step S153.

[0108] (Free energy calculation process) (Step S153) The conformation generation unit 315 generates one initial conformation for a candidate reaction based on the structure of the generated candidate reaction. For example, the conformation generation unit 315 generates one initial conformation using a trained model such as a classical force field or NNP based on the structure of the candidate reaction, i.e., the initial structure ARS after the reaction. The conformation generation unit 315 stores the generated initial conformation in the main memory 33 or auxiliary memory 35. In this application example, since multiple conformations are generated by the extended ensemble molecular dynamics simulation, one initial conformation is sufficient. However, multiple initial conformations may be generated, as in the embodiment.

[0109] (Step S154) The reaction parameter calculation unit 317 determines the free energy surface for the reaction candidate by performing an extended ensemble molecular dynamics simulation based on a single initial conformation, accompanied by structural optimization using a trained model. The free energy corresponds to, for example, Gibbs free energy. Specifically, the extended ensemble molecular dynamics simulation is performed using the metadynamics method. Alternatively, the extended ensemble molecular dynamics simulation may be performed using umbrella sampling.

[0110] For example, the reaction parameter calculation unit 317 adds a potential represented by a Gaussian distribution (hereinafter referred to as the Gaussian potential) to the conformation being calculated at each of the multiple calculation steps in the polymer formation process by the extended ensemble molecular dynamics simulation. The addition of the Gaussian potential corresponds, for example, to the transfer of thermal energy to the conformation. The magnitude of the added Gaussian potential may be appropriately adjusted in relation to the current potential level in the extended ensemble molecular dynamics simulation. Next, the reaction parameter calculation unit 317 performs the extended ensemble molecular dynamics simulation on the state in which the Gaussian potential has been added. The details of the execution of the extended ensemble molecular dynamics simulation are known, so a description will be omitted.

[0111] Next, the reaction parameter calculation unit 317 inputs the conformation after the execution of the extended ensemble molecular dynamics simulation into the trained model to obtain the energy corresponding to that conformation. Subsequently, the reaction parameter calculation unit 317 calculates the force related to that conformation using the obtained energy. The reaction parameter calculation unit 317 then performs structural optimization with respect to that conformation using the calculated force.

[0112] The reaction parameter calculation unit 317 adds a Gaussian potential to the structurally optimized conformation and performs the extended ensemble molecular dynamics simulation again. The reaction parameter calculation unit 317 repeats the above calculation until the calculated force is less than or equal to a predetermined force and / or the calculated energy is less than or equal to a predetermined energy. The termination condition for the extended ensemble molecular dynamics simulation is not limited to the above, and may be, for example, that the structurally optimized conformation satisfies a predetermined condition, as shown in step S125.

[0113] Figure 16 shows an example of an extended ensemble molecular dynamics simulation using metadynamics. In the graph shown in Figure 16, the horizontal axis represents CV and the vertical axis represents free energy. The horizontal axis may also be represented by interatomic distance. In the graph, FES represents the free energy surface. CF represents the conformation of the reacting compound. As shown in Figure 16, in the initial MDS of the extended ensemble molecular dynamics simulation, the initial conformational CF is set at a position with a relatively high potential. As shown in Figure 16, in the first state MD1, a Gaussian potential is appropriately added, and the energy of conformational CF, CFE, increases.

[0114] When the energy CFE of conformation CF rises to the activation barrier AB due to the addition of the Gaussian potential (MD2), the energy CFE of the reacting compound may exceed the activation barrier AB, as shown in the extended ensemble molecular dynamics simulation (MD3). At this point, the Gaussian potential AP is added while the energy CFE exceeds the activation barrier AB. When the energy CFE of conformation CF rises to the activation barrier AB due to the addition of the Gaussian potential (MD4), the reacting compound becomes capable of transitioning between the pre- and post-reaction states, as shown in the extended ensemble molecular dynamics simulation (MDE). At this point, the extended ensemble molecular dynamics simulation terminates.

[0115] The reaction parameter calculation unit 317 determines the free energy surface FES based on the Gaussian potential added by the extended ensemble molecular dynamics simulation. Specifically, the reaction parameter calculation unit 317 determines the free energy surface FES by inverting the free energy of the added Gaussian potential CFE in the final state MDE obtained by the extended ensemble molecular dynamics simulation, with the horizontal axis as the rotation axis.

[0116] (Step S155) The reaction parameter calculation unit 317 calculates the free energy based on the free energy surface. For example, the reaction parameter calculation unit 317 calculates the free energy by analyzing the free energy surface FES. Since known methods can be applied to this analysis, a detailed explanation is omitted. The reaction parameter calculation unit 317 stores the calculated free energy in the main memory 33 or auxiliary memory 35 in association with the reaction.

[0117] (Step S156) If the free energy calculations for all reaction candidates have been completed (Yes in step S156), the process in step S157 is executed. If the free energy calculations for all reaction candidates have not been completed (No in step S156), the process from step S153 onwards is repeated for the reaction candidates for which the free energy calculations are incomplete.

[0118] (Step S157) The reaction parameter calculation unit 317 generates a list of free energies corresponding to the reaction list. For example, the reaction parameter calculation unit 317 generates a list of free energies corresponding to multiple reactions in the reaction list RL. The reaction parameter calculation unit 317 stores the generated list of free energies in the main memory 33 or the auxiliary memory 35.

[0119] The free energy calculation process performed by the information processing device 1 has been described above. The following describes the process of calculating (estimating) the primary structure, which is performed by the information processing device 1. As an example, the primary structure calculation will be explained using the monomer conversion rate, polymer molecular weight, and polymer molecular weight distribution as examples.

[0120] Figure 17 is a flowchart showing an example of the procedure for estimating the monomer conversion rate, polymer molecular weight, and polymer molecular weight distribution as the primary structure of a polymer (hereinafter referred to as the molecular weight estimation process).

[0121] (Molecular weight estimation process) (Step S171) A list of multiple elementary reactions is input via an input device. Elementary reactions are, for example, the individual basic reactions that make up a complex chemical reaction. This step corresponds to the input of the reaction list in Figure 15.

[0122] (Step S172) In response to the input of the list of elementary reactions, the free energy calculation process shown in FIG. 15 is executed. As a result, a list of free energies of all reactions in the input list of elementary reactions is generated.

[0123] (Step S173) The primary structure calculation unit 319 calculates the reaction rate constant for each of the plurality of elementary reactions using the free energy for each of the plurality of elementary reactions. The reaction rate constant k is expressed by the formula k = { (k b × T) / h} × exp { -G / (RT)} using Planck's constant h, the temperature T at which the reaction occurs, the gas constant R, the Boltzmann constant k b and the free energy G for each elementary reaction. In the reaction rate constant k, the coefficient { (k b × T) / h} corresponds to the frequency factor. That is, the reaction rate constant k is the frequency factor { (k b × T) / h} multiplied by the exponential of the activation free energy / (temperature × gas constant). Note that the reaction rate constant k and the frequency factor { (k b × T) / h} may be calculated by the reaction parameter calculation unit 317.

[0124] For example, if the free energy G <000 , the primary structure calculation unit 319 calculates the reaction rate constant k I d = { (k b × T) / h} × exp { -G I / (RT)} corresponding to the initiation reaction. Also, if the free energy G P is generated in the propagation reaction, the primary structure calculation unit 319 calculates the reaction rate constant k P = { (k b × T) / h} × exp { -G P / (RT)} corresponding to the propagation reaction. Also, if the free energy G T is generated in the transfer reaction, the primary structure calculation unit 319 calculates the reaction rate constant k T = { (k b×T) / h}×exp{-G T Calculate / (RT)}.

[0125] (Step S174) The primary structural calculation unit 319 solves a system of differential equations relating to mass balance and moment balance using the reaction rate constants of multiple elementary reactions. Known equations can be used as the system of differential equations relating to mass balance and moment balance, and are pre-set and stored in the main memory 33 or auxiliary memory 35. Furthermore, the method for solving the system of differential equations using reaction rate constants is known, so its explanation is omitted. Note that some of the multiple reaction rate constants corresponding to multiple elementary reactions may be experimental values.

[0126] (Step S175) The primary structure calculation unit 319 uses the solution to simultaneous differential equations to calculate the monomer conversion rate, the molecular weight of the polymer (number-average molecular weight and weight-average molecular weight, etc.), and the distribution of the polymer's molecular weight. Since the calculation methods for the monomer conversion rate, polymer molecular weight, and distribution of the polymer's molecular weight are known, a detailed explanation is omitted. Thus, by taking the elementary reactions and reaction rate constants to be used as the calculation target, and the initial monomer concentration as input, and executing the molecular weight estimation process, the monomer conversion rate and molecular weight distribution can be estimated. For example, a molecular weight simulator with this molecular weight estimation process can be applied to major chain polymerization reactions and Normal ATRP (Atom Transfer Radical Polymerization) and RTCP (Reversible Chain Transfer Catalyzed Polymerization).

[0127] Based on the above, the information processing device 1 according to the first application example of this embodiment determines the free energy surface for a candidate reaction by performing an extended ensemble molecular dynamics simulation based on the initial conformation, and calculates the activation free energy by analyzing the determined free energy surface. That is, when calculating free energy as a reaction parameter, the information processing device 1 according to the first application example of this embodiment generates one initial conformation for the candidate reaction based on the structure of the candidate reaction, and determines the free energy surface for the candidate reaction by performing an extended ensemble molecular dynamics simulation for the one initial conformation with structural optimization using a trained model, and calculates the free energy by analyzing the determined free energy surface.

[0128] As a result, the information processing device 1 according to this embodiment can calculate the primary structure of a polymer by using a trained model such as NNP, along with structural optimization in the generation process, and then using the trained model and the initial conformation to calculate the free energy. As a result, the information processing device 1 according to this embodiment can calculate the free energy with lower computational cost compared to DFT and with high accuracy due to structural optimization by the trained model, without conducting experiments. Other effects are the same as in the embodiment, so their explanation will be omitted.

[0129] (Second application example) This application example involves generating multiple reaction parameters using the following procedure, as described in the embodiment, and storing the generated reaction parameters as a database. First, the acquisition unit 311 acquires information on multiple reactions related to polymer formation. Next, the conformation generation unit 315 generates at least one initial conformation for multiple reaction candidates corresponding to the multiple reactions based on the information on the multiple reactions. Subsequently, the reaction parameter calculation unit 317 calculates at least one reaction parameter for each of the multiple reactions based on the generated initial conformation and machine learning potential. Finally, the reaction parameter calculation unit 317 stores the calculated at least one reaction parameter as a database in the main memory 33 and / or auxiliary memory 35. The above series of processes may be performed for different polymers. In this case, the reaction parameter calculation unit 317 may store the reaction parameters calculated according to the different polymers as a database in the main memory 33 and / or auxiliary memory 35.

[0130] The acquisition unit 311 acquires at least one reaction parameter from a plurality of reaction parameters stored in memory. The primary structure calculation unit 319 uses the reaction parameters stored in the database, i.e., the acquired reaction parameters, to calculate the primary structure of the polymer desired by the user. The calculation of the primary structure is described in accordance with the embodiment and the first application example, so the explanation is omitted. Note that the information processing device that calculates the reaction parameters stored in the database and the information processing device that performs the calculation of the primary structure using the database may be the same device or different devices.

[0131] From the above, the information processing device 1 according to the second application example of this embodiment obtains at least one reaction parameter from a plurality of reaction parameters stored in one memory, calculates the primary structure of the polymer using the obtained reaction parameter, and each of the plurality of reaction parameters stored in one memory in the information processing device 1 is generated by obtaining information on a plurality of reactions related to the formation of the polymer, generating at least one initial conformation for a plurality of reaction candidates corresponding to the plurality of reactions based on the information on the plurality of reactions, and calculating at least one reaction parameter for each of the plurality of reactions based on the initial conformation and machine learning potential.

[0132] Based on these considerations, the information processing device 1 according to the second application example can pre-calculate reaction parameters according to the polymer and store them as a database in the main memory 33 or auxiliary memory 35, thereby reducing the computational cost of primary structure calculations such as stereoregularity estimation and molecular weight estimation. Other effects are the same as those in the embodiment and the first application example, so their explanation will be omitted.

[0133] When the technical concept in the embodiment is realized by an information processing method, the information processing method involves at least one computer acquiring information on multiple reactions related to polymer formation, generating at least one initial conformation for multiple candidate reactions corresponding to the multiple reactions based on the information on the multiple reactions, calculating at least one reaction parameter for each of the multiple reactions based on the initial conformation and machine learning potential, and calculating the primary structure of the polymer using the reaction parameter. The procedures and effects of various processes such as activation energy calculation, free energy calculation, stereoregularity estimation, and molecular weight estimation in the information processing method are the same as those described in the embodiment, the first application example, and the second application example, so a detailed explanation is omitted.

[0134] When the technical concept in the embodiment is realized by an information processing program, the information processing program enables at least one computer to acquire information on multiple reactions related to polymer formation, generate at least one initial conformation for multiple candidate reactions corresponding to the multiple reactions based on the information on the multiple reactions, calculate at least one reaction parameter for each of the multiple reactions based on the initial conformation and machine learning potential, and calculate the primary structure of the polymer using the reaction parameter.

[0135] For example, the information processing program can also be implemented by installing it on a computer in a simulation device or simulation server capable of calculating monomer sequences, and then loading it into memory. In this case, the program that can cause the computer to execute the MD simulation can also be stored and distributed on storage media such as magnetic disks (hard disks, etc.), optical disks (CD-ROMs, DVDs, etc.), and semiconductor memory. The procedures and effects of various processes such as activation energy calculation, free energy calculation, stereoregularity estimation, and molecular weight estimation by the information processing program are the same as those described in the embodiments, the first application example, and the second application example, so a detailed explanation is omitted.

[0136] In the embodiments described above, some or all of the devices may be composed of hardware, or they may be composed of information processing by software (programs) executed by a CPU or GPU. If the information processing is composed of software, the software that realizes at least some of the functions of the devices in the embodiments described above may be stored on a non-temporary storage medium (non-temporary computer-readable medium) such as a flexible disk, CD-ROM (Compact Disc-Read Only Memory), or USB memory, and the information processing of the software may be executed by having the computer 30 read it. Alternatively, the software may be downloaded via a communication network 5. Furthermore, the information processing may be executed by hardware by implementing the software on a circuit such as an ASIC or FPGA.

[0137] The type of storage medium used to store the software is not limited. The storage medium is not limited to removable media such as magnetic disks or optical disks; it may also be a fixed storage medium such as a hard disk or memory. Furthermore, the storage medium may be located inside or outside the computer.

[0138] Where the expression "at least one of a, b, and c" or "at least one of a, b, or c" (including similar expressions) is used in this specification (including the claims), it includes any of a, b, c, ab, ac, bc, or abc. It also includes multiple instances of any element, such as aa, abb, aabbcc, etc. Furthermore, it includes adding other elements other than the enumerated elements (a, b, and c), such as abcd which has d.

[0139] In this specification (including the claims), when expressions such as "data as input / based on / according to / in accordance with data" (including similar expressions) are used, unless otherwise specified, this includes cases where the data itself is used as input, or where the data has been processed in some way (e.g., data with added noise, normalized data, intermediate representations of the data, etc.) is used as input. Furthermore, when it is stated that some result is obtained "based on / according to / in accordance with data", this includes cases where the result is obtained based solely on the data in question, as well as cases where the result is also influenced by other data, factors, conditions, and / or states other than the data in question. Furthermore, when it is stated that "data is output", unless otherwise specified, this includes cases where the data itself is used as output, or where the data has been processed in some way (e.g., data with added noise, normalized data, intermediate representations of the data, etc.) is used as output.

[0140] In this specification (including the claims), the terms “connected” and “coupled” are intended to be non-restrictive terms that include any direct connection / coupling, indirect connection / coupling, electrical connection / coupling, communicative connection / coupling, operational connection / coupling, physical connection / coupling, etc. The terms should be interpreted as appropriate in the context in which they are used, but any form of connection / coupling that is not intentionally or naturally excluded should be interpreted non-restrictively as being included in the terms.

[0141] In this specification (including the claims), when the expression "A configured to B" is used, it may include that the physical structure of element A has a configuration capable of performing operation B, and that the permanent or temporary setting / configuration of element A is configured to actually perform operation B. For example, if element A is a general-purpose processor, it is sufficient that the processor has a hardware configuration capable of performing operation B, and that it is configured to actually perform operation B by the setting of a permanent or temporary program (instruction). Furthermore, if element A is a dedicated processor or dedicated arithmetic circuit, it is sufficient that the circuit structure of the processor is implemented to actually perform operation B, regardless of whether control instructions and data are actually attached.

[0142] Wherever terms meaning "comprising" or "possessing" (e.g., "comprising / including" and "having") are used in this specification (including the claims), they are intended to be open-ended terms, including cases where the subject matter of such terms is not the object of the term. Where the object of such terms meaning "comprising" or "possessing" is an expression that does not specify a quantity or suggests a singular number (an expression with the article "a" or "an"), such expression should be interpreted as not being limited to a specific number.

[0143] In this specification (including the claims), even if expressions such as "one or more" or "at least one" are used in one place, and expressions that do not specify a quantity or suggest a singularity (expressions using the articles a or an) are used in another place, the latter expressions are not intended to mean "one." In general, expressions that do not specify a quantity or suggest a singularity (expressions using the articles a or an) should be interpreted as not necessarily being limited to a specific number.

[0144] In this specification, if a particular configuration of an embodiment is described as yielding a specific advantage or result, it should be understood, unless otherwise stated, that the same advantage or result can also be obtained from one or more other embodiments having the same configuration. However, it should be understood that the presence or absence of such an advantage or result generally depends on various factors, conditions, and / or states, and that the configuration does not necessarily guarantee that the advantage or result can be obtained. The advantage or result can only be obtained from the configuration described in the embodiment when various factors, conditions, and / or states are met, and the advantage or result cannot necessarily be obtained in the invention claimed to define that configuration or a similar configuration.

[0145] In this specification (including the claims), when terms such as "maximize" are used, they include finding the global maximum value, finding an approximation of the global maximum value, finding the local maximum value, and finding an approximation of the local maximum value, and should be interpreted appropriately depending on the context in which the term is used. They also include finding approximations of these maximum values ​​probabilistically or heuristically. Similarly, when terms such as "minimize" are used, they include finding the global minimum value, finding an approximation of the global minimum value, finding the local minimum value, and finding an approximation of the local minimum value, and should be interpreted appropriately depending on the context in which the term is used. They also include finding approximations of these minimum values ​​probabilistically or heuristically. Similarly, when terms such as "optimize" are used, they include finding the global optimal value, finding an approximation of the global optimal value, finding the local optimal value, and finding an approximation of the local optimal value, and should be interpreted appropriately depending on the context in which the term is used. They also include finding approximations of these optimal values ​​probabilistically or heuristically.

[0146] In this specification (including the claims), when multiple hardware components perform a predetermined process, each component may cooperate to perform the predetermined process, or some components may perform all of the predetermined process. Alternatively, some components may perform part of the predetermined process, while other components perform the remainder. In this specification (including the claims), when expressions such as "one or more hardware components perform a first process, and the one or more hardware components perform a second process" are used, the hardware component performing the first process and the hardware component performing the second process may be the same or different. In other words, it is sufficient that the hardware component performing the first process and the hardware component performing the second process are included in the one or more hardware components. Hardware may include electronic circuits or devices containing electronic circuits.

[0147] In this specification (including the claims), when multiple memory devices store data, each of the multiple memory devices may store only a portion of the data or the entire data.

[0148] While embodiments of this disclosure have been described in detail above, this disclosure is not limited to the individual embodiments described above. Various additions, modifications, substitutions, and partial deletions are possible, provided that they do not depart from the conceptual idea and spirit of the present invention derived from the claims and their equivalents. For example, where numerical values ​​or mathematical formulas are used in the description in all of the embodiments described above, they are provided as examples only and are not limited thereto. Also, the order of operations in the embodiments is provided as examples only and is not limited thereto. [Explanation of symbols]

[0149] 1. Information Processing Device 5. Communication Network 9A external device 9B External device 30 Computers 31 processors 33 Main memory 35 Auxiliary storage device 37 Network Interfaces 39 Device Interfaces 41 Bus 311 Acquisition Department 313 Structure generation part 315 Conformation generator 317 Reaction parameter calculation unit 319 Primary structure calculation section

Claims

1. At least one memory, It comprises at least one processor, The aforementioned at least one processor is We obtained information on multiple reactions related to polymer formation, Based on the information regarding the plurality of reactions, at least one initial conformation is generated for a plurality of candidate reactions corresponding to the plurality of reactions. Based on the initial conformation and the machine learning potential, at least one reaction parameter for each of the multiple reactions is calculated. The primary structure of the polymer is calculated using the aforementioned reaction parameters. Information processing device.

2. The aforementioned at least one processor is Based on the information regarding the plurality of reactions, the structures of candidate products for the plurality of reactions are generated. Based on the structure of the candidate object, at least one initial conformation of the candidate object is generated. The information processing apparatus according to claim 1.

3. The aforementioned reaction parameters include at least one of the following: activation energy, activation free energy, frequency factor, and reaction rate constant. The information processing apparatus according to claim 1.

4. When calculating the activation energy as the reaction parameter, the at least one processor, Using the initial conformation, multiple conformations before the reaction are generated based on structural optimization using the machine learning potential. The activation energy is calculated by performing a transition state search using the machine learning potential with the multiple conformations prior to the reaction. The information processing apparatus according to claim 3.

5. The aforementioned at least one processor is By performing the structural optimization while changing the bond length, multiple conformations before the reaction are generated. The information processing apparatus according to claim 4.

6. When calculating the activation energy as the reaction parameter, the at least one processor, Using the aforementioned activation energy, the reaction probability between the growth radical or initiator and the monomer in the generation of the candidate product is calculated. The information processing apparatus according to claim 3.

7. When calculating the activation free energy as the reaction parameter, at least one processor, By performing extended ensemble molecular dynamics simulations based on the aforementioned initial conformation, the free energy surface for the candidate reaction is determined. The activation free energy is calculated by analyzing the aforementioned free energy surface. The information processing apparatus according to claim 3.

8. The aforementioned machine learning potential is the neural network potential. The information processing apparatus according to any one of claims 1 to 7.

9. The initial conformation is the initial structure after the reaction, and is three-dimensional data of multiple atoms constituting the candidate product of the reaction. The information processing apparatus according to any one of claims 1 to 7.

10. At least one memory, It comprises at least one processor, The aforementioned at least one processor is From the multiple reaction parameters stored in the aforementioned single memory, at least one reaction parameter is obtained, Using the aforementioned reaction parameters, the primary structure of the polymer is calculated. Each of the multiple reaction parameters stored in the aforementioned single memory is: We obtained information on multiple reactions related to polymer formation, Based on the information regarding the plurality of reactions, at least one initial conformation is generated for a plurality of candidate reactions corresponding to the plurality of reactions. This is generated by calculating at least one reaction parameter for each of the multiple reactions based on the initial conformation and the machine learning potential. Information processing device.

11. The primary structure includes at least one of the following: monomer chain distribution, head-to-tail bonding, stereoregularity, and molecular weight distribution. The information processing apparatus according to claim 1 or claim 10.

12. At least one computer, We obtained information on multiple reactions related to polymer formation, Based on the information regarding the plurality of reactions, at least one initial conformation is generated for a plurality of candidate reactions corresponding to the plurality of reactions. Based on the initial conformation and the machine learning potential, at least one reaction parameter for each of the multiple reactions is calculated. The primary structure of the polymer is calculated using the aforementioned reaction parameters. Information processing methods.

13. On at least one computer, We obtained information on multiple reactions related to polymer formation, Based on the information regarding the plurality of reactions, at least one initial conformation is generated for a plurality of candidate reactions corresponding to the plurality of reactions. Based on the initial conformation and the machine learning potential, at least one reaction parameter for each of the multiple reactions is calculated. The primary structure of the polymer is calculated using the aforementioned reaction parameters. An information processing program that makes this possible.

Images