Photosensitive resin composition, film-shaped resin, and resin sheet

Abstract

Disclosed is a photosensitive resin composition which contains a polymer composed of a plurality of kinds of monomers, wherein: the dispersion term of the Hansen solubility parameter as calculated on the basis of the monomers is 16.99 to 17.35 inclusive; the number of carboxylic acids in each molecule as calculated on the basis of the monomers is 0.272 to 0.304 inclusive; the logarithmic value of the octanol / water partition coefficient as calculated on the basis of the monomers is 2.16 to 2.79 inclusive; and the glass transition temperature as calculated on the basis of the monomers is 354 K to 382 K inclusive.

Description

Photosensitive resin composition, film-like resin, and resin sheet

[0001] The present disclosure relates to a photosensitive resin composition, a film-like resin, and a resin sheet.

[0002] Photosensitive resin sheets (also called "dry film resists") are used to form wiring on substrates such as semiconductor wafers (see, for example, Patent Document 1). In recent years, with the trend toward finer wiring on semiconductor package substrates, there has been a demand for dry film resists that offer both high resolution and high adhesion.

[0003] Japanese Patent Application Laid-Open No. 2013-216804

[0004] However, conventional dry film resists have insufficient polymer resolution or adhesion, making it difficult to form fine wiring.

[0005] One aspect of the present disclosure provides a photosensitive resin composition, a film-like resin, and a resin sheet that are excellent in resolution and adhesion.

[0006] The present disclosure has the following configuration.

[0007] <1> A photosensitive resin composition containing a polymer constituted by a plurality of types of monomers, wherein the dispersion term of the Hansen solubility parameter calculated based on the monomers is 16.99 or more and 17.35 or less, the number of carboxylic acids in the molecule calculated based on the monomers is 0.272 or more and 0.304 or less, the logarithm of the octanol / water partition coefficient calculated based on the monomers is 2.16 or more and 2.79 or less, and the glass transition temperature calculated based on the monomers is 354 K or more and 382 K or less.

[0008] <2> The photosensitive resin composition according to <1> above, wherein the dispersion term of the Hansen solubility parameter is 17.04 or more and 17.19 or less, the number of carboxylic acids is 0.273 or more and 0.290 or less, the logarithm of the octanol / water partition coefficient is 2.29 or more and 2.78 or less, and the glass transition temperature is 365 K or more and 382 K or less.

[0009] <3> The photosensitive resin composition according to <1> or <2>, wherein the monomer contains acrylic acid and styrene.

[0010] <4> The photosensitive resin composition according to <1> above, wherein the monomer includes a third component different from both acrylic acid and styrene, and the third component is 3-phenoxybenzyl acrylate, 3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate, octadecyl 2-methylprop-2-enoate, 4-hydroxyphenyl 2-methyl-2-propenoate, octadecyl 2-methylprop-2-enoate, dicyclopentanyl methacrylate, decyl prop-2-enoate, dodecyl acrylate, or isooctadecyl 2-propenoate.

[0011] <5> The monomer includes a third component different from both acrylic acid and styrene, and the third component is 10-hydroxy-2,2,4-trimethyl-4-[(trimethylsilyl)oxy]-3,8-dioxa-2,4-disilaundecan-11-yl 2-methylprop-2-enoate, hexadecyl prop-2-enoate, docosylprop-2-enoate, 1-ethylcyclohexyl 2-methylprop-2-enoate, docosylprop-2-enoate, tetradecyl prop-2-enoate, adamantan-1-yl 2-methylprop-2-enoate, 3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate, or 2-propenoic acid isooctadecyl ester. The photosensitive resin composition according to the above item <2>.

[0012] <6> The photosensitive resin composition according to <4> or <5>, wherein the polymer contains the third component in an amount of 13% by weight or more and 22% by weight or less.

[0013] <7> A film-like resin obtained by forming the photosensitive resin composition according to any one of <1> to <6> above onto a film.

[0014] <8> A resin sheet comprising: a substrate; and a resin layer formed on the substrate and made of the photosensitive resin composition according to any one of <1> to <6>.

[0015] According to one aspect of the present disclosure, it is possible to provide a photosensitive resin composition, a film-like resin, and a resin sheet that are excellent in resolution and adhesion.

[0016] FIG. 1 is a diagram showing extraction results of Example 1. FIG. 2 is a diagram showing extraction results of Example 2. FIG. 3 is a schematic diagram showing an example of a film-like resin. FIG. 4 is a schematic diagram showing an example of a resin sheet. FIG. 5 is a block diagram showing an example of the overall configuration of a composition proposal system. FIG. 6 is a block diagram showing an example of the hardware configuration of a computer. FIG. 7 is a block diagram showing an example of the functional configuration of the composition proposal system. FIG. 8 is a diagram showing an example of experimental data. FIG. 9 is a diagram showing an example of descriptor data. FIG. 10 is a block diagram showing an example of an adhesion prediction model. FIG. 11 is a graph showing an example of the accuracy of the adhesion prediction model. FIG. 12 is a block diagram showing an example of a resolution prediction model. FIG. 13 is a graph showing an example of the accuracy of the resolution prediction model. FIG. 14 is a flowchart showing an example of a descriptor calculation process. FIG. 15 is a flowchart showing an example of a model learning process. FIG. 16 is a flowchart showing an example of a composition proposal process.

[0017] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0018] [Photosensitive Resin Composition] The photosensitive resin composition in this embodiment is a composition used to produce a dry film resist. The photosensitive resin composition contains a polymer composed of multiple types of monomers. The monomers that constitute the polymer include at least acrylic acid (AA) and styrene (STC). Hereinafter, the polymer contained in the photosensitive resin composition will be referred to as a "base polymer."

[0019] The photosensitive resin composition may contain, in addition to the base polymer, a (meth)acrylate compound, an epoxy resin, a photoinitiator, etc. The monomer constituting the base polymer may contain a third component other than both acrylic acid and styrene.

[0020] In this embodiment, the blending amount of acrylic acid is 18% by weight or more and 26% by weight or less. The blending amount of styrene is 60% by weight or more and 80% by weight or less. The blending amount of the third component is the portion of the base polymer excluding acrylic acid and styrene. In other words, the blending amount of the third component is (100 - (blended amount of acrylic acid + blended amount of styrene)) [wt %]. The blending amount of the third component is preferably 13% by weight or more and 22% by weight or less.

[0021] <Evaluation Index> The dry film resist produced using the photosensitive resin composition of this embodiment achieves both high resolution and high adhesion. In this embodiment, the evaluation index for resolution is "resolution 3x / x". The evaluation index for adhesion is "16 μm pitch adhesion".

[0022] Resolution 3x / x is the minimum value of x at which a pattern can be formed when a fine line pattern is formed in dry film resist, where 3x is the line width and x is the space width between the lines. Resolution 3x / x can be evaluated as having a higher resolution as the value of x decreases. Whether a pattern can be formed can be confirmed by observing the created pattern using a microscope to check for residues or resist collapse.

[0023] The 16 μm pitch adhesion is the minimum line width at which a fine line pattern can be formed in a dry film resist when the space width is widened in an evaluation pattern with a line pitch of 16 μm. The smaller the line width value, the higher the adhesion can be evaluated. The ability to form a pattern can be confirmed by observing the created pattern using a microscope to check for residue or resist collapse.

[0024] The 16 μm pitch adhesion may also be offset pattern adhesion. The offset pattern adhesion is the minimum line width that can be formed when a fine line pattern is formed in a dry film resist and the total line width and space width is 16 μm, and the space width is widened in an evaluation pattern. The smaller the offset pattern adhesion value, the higher the adhesion can be evaluated.

[0025] <Parameters> In the photosensitive resin composition of this embodiment, four parameters calculated based on the monomers constituting the base polymer fall within predetermined ranges. The first parameter is the dispersion term of the Hansen solubility parameter (hereinafter also referred to as "HSP_δD"). The second parameter is the number of carboxylic acids in the molecule (hereinafter also referred to simply as "number of carboxylic acids"). The third parameter is the logarithm of the octanol / water partition coefficient (hereinafter also referred to as "logP"). The fourth parameter is the glass transition temperature (hereinafter also referred to as "polymer Tg").

[0026] <<Dispersion term of Hansen solubility parameter (HSP_δD)>> The Hansen solubility parameter represents the strength of interaction between polymers, and is therefore thought to affect resolution or adhesion. Of the Hansen solubility parameters, the dispersion term δD is used instead of the polar term δP or the hydrogen bond term δH because it is thought to represent the strength of interaction between the benzene rings of styrene.

[0027] The HSP_δD of the base polymer is calculated by weighting and adding the HSP_δD of each monomer constituting the base polymer by molar ratio. Specifically, the HSP_δD of the base polymer can be calculated using formula (1).

[0028]

[0029] where X is the HSP_δD of the base polymer, and x ｉ is the HSP_δD of the i-th monomer, and n ｉ is the molar ratio of the i-th monomer.

[0030] The HSP_δD of a monomer can be calculated using, for example, HSPiP, which is software specialized for calculating Hansen solubility parameters.

[0031] The HSP_δD of the base polymer is preferably 16.99 or greater and 17.35 or less, and more preferably 17.04 or greater and 17.19 or less.

[0032] <Number of carboxylic acids in a molecule> In dry film resists, carboxylic acids react with bases during development and peeling, so the number of carboxylic acids in a molecule is thought to affect resolution and adhesion.

[0033] The number of carboxylic acids in the base polymer is calculated by weighting the number of carboxylic acids in each monomer constituting the base polymer by the molar ratio and adding them together. Specifically, the number of carboxylic acids in the base polymer can be calculated using the above formula (1).

[0034] In the formula (1), X is the number of carboxylic acids in the base polymer, and x ｉ is the number of carboxylic acids in the i-th monomer, and n ｉ is the molar ratio of the i-th monomer.

[0035] The number of carboxylic acids in a monomer can be calculated using, for example, RDKit, an open-source library widely used in the field of cheminformatics.

[0036] The number of carboxylic acids in the base polymer is preferably 0.272 or more and 0.304 or less, and more preferably 0.273 or more and 0.290 or less.

[0037] <Logarithm of Octanol / Water Partition Coefficient (logP)> Dry film resists use aqueous solutions for development. Therefore, the hydrophilicity of the base polymer is thought to affect resolution or adhesion. In addition, an index showing hydrophilicity can also be thought of as an index of the interaction between polymers.

[0038] The log P of the base polymer is calculated by weighting and adding the log P of each monomer constituting the base polymer by the molar ratio. Specifically, the log P of the base polymer can be calculated by the above formula (1).

[0039] In formula (1), X is the log P of the base polymer, and x ｉ is the log P of the i-th monomer, and n ｉ is the molar ratio of the i-th monomer.

[0040] The log P of a monomer can be calculated using, for example, COSMOtherm (registered trademark). COSMOtherm is thermodynamic property prediction software that uses the COSMO-RS method. The COSMO-RS method is a technique for predicting the physical properties of chemical substances based on the surface charge of the molecules obtained by quantum chemical calculations.

[0041] The log P of the base polymer is preferably 2.16 or more and 2.79 or less, and more preferably 2.29 or more and 2.78 or less.

[0042] Glass Transition Temperature (Polymer Tg) Polymer Tg indicates the mobility of a polymer chain. Therefore, it is believed that the polymer Tg of the base polymer affects the resolution or adhesion.

[0043] The polymer Tg of the base polymer is calculated by the Fox equation using the polymer Tg of a homopolymer composed of each monomer that constitutes the base polymer. Specifically, the polymer Tg of the base polymer can be calculated by equation (2).

[0044]

[0045] where TG is the polymer Tg of the copolymer (i.e., the base polymer), and TG ｉ is the polymer Tg of the homopolymer composed of the i-th monomer, and n ｉ is the molar ratio of the i-th monomer.

[0046] The polymer Tg of the homopolymer may be a literature value or a predicted value using a trained machine learning model.

[0047] The polymer Tg of the base polymer is preferably 354K or more and 382K or less, and more preferably 365K or more and 382K or less.

[0048] <Extraction Method> The photosensitive resin composition in this embodiment is a candidate composition that satisfies both the target values ​​for resolution 3x / x and 16 μm pitch adhesion, extracted from a large number of candidate compositions produced under specified production conditions. In this embodiment, the resolution 3x / x and 16 μm pitch adhesion are predicted using a trained machine learning model. The machine learning model and the prediction method using the machine learning model will be described later.

[0049] In this embodiment, the production conditions are defined as follows: The blending amount of acrylic acid is generated in 1 wt% increments within the range of 18 wt% to 26 wt%. That is, the blending amount options for acrylic acid are 18 wt%, 19 wt%, 20 wt%, ... 26 wt%. The blending amount of styrene is generated in 2 wt% increments within the range of 60 wt% to 80 wt%. That is, the blending amount options for styrene are 60 wt%, 62 wt%, 64 wt%, ... 80 wt%. The sum of the blending amounts of each monomer is 100 wt%. That is, the blending amount of the third component is (100 - (blended amount of acrylic acid + blended amount of styrene)) [wt%]. The third component is selected from a selection of monomers prepared in advance. For example, approximately 200 types of monomers may be included in the options.

[0050] Candidate compositions are generated as follows. First, all combinations of the acrylic acid content options and the styrene content options are generated. Next, for each combination of the content options, the content of the third component is calculated. Then, for each combination of the content options, all of the monomer options are combined as the third component. This generates all candidate compositions.

[0051] Example 1 In Example 1, target properties were set to 16 μm pitch adhesion of less than 4.7 and resolution 3x / x of less than 7, and candidate compositions were extracted whose predicted values, which satisfied the target properties, were predicted using a trained machine learning model.

[0052] For base polymers contained in conventionally known photosensitive resin compositions, the best 16 μm pitch adhesion value was 4.7. Therefore, the target value for 16 μm pitch adhesion was set to 4.7. Furthermore, for conventional base polymers, the best resolution 3x / x value was 5.5, but a resolution 3x / x of 7 is a sufficiently high standard for the performance required of a dry film resist.

[0053] Figure 1 shows the extraction results of Example 1. As shown in Figure 1, the candidate compositions that satisfy the above target properties have each parameter falling within the following ranges: HSP_δD is 16.99 or more and 17.35 or less; the number of carboxylic acids is 0.272 or more and 0.304 or less; log P is 2.16 or more and 2.79 or less; and the polymer Tg is 354 K or more and 382 K or less.

[0054] Furthermore, the candidate compositions extracted in Example 1 contain the third component in an amount of 13% by weight or more and 22% by weight or less.

[0055] Furthermore, the candidate compositions extracted in Example 1 have a third component that is one of the following:・3-phenoxybenzyl acrylate ・3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate ・Octadecyl 2-methylprop-2-enoate ・4-hydroxyphenyl 2-methyl-2-propenoate (PQMA) ・Octadecyl 2-methylprop-2-enoate 2-methylprop-2-enoate, dicyclopentanyl methacrylate (TCDMA), decyl prop-2-enoate, dodecyl acrylate (LA), and 2-propenoic acid isooctadecyl ester (ISTA). Example 2 In Example 2, target properties were set to a 16 μm pitch adhesion of less than 4.7 and a resolution 3x / x of less than 6, and candidate compositions were extracted whose predicted values ​​satisfied the target properties using a trained machine learning model. That is, in Example 2, candidate compositions exhibiting higher resolution than Example 1 were extracted.

[0056] Figure 2 shows the extraction results of Example 2. As shown in Figure 2, the candidate compositions that satisfy the above target properties have each parameter falling within the following ranges: HSP_δD is 17.04 or more and 17.19 or less. The number of carboxylic acids is 0.273 or more and 0.290 or less. Log P is 2.29 or more and 2.78 or less. Polymer Tg is 365K or more and 382K or less.

[0057] Furthermore, the candidate compositions extracted in Example 2 contain the third component in an amount of 13% by weight or more and 22% by weight or less.

[0058] Furthermore, the candidate compositions extracted in Example 2 have a third component that is one of the following:・10-hydroxy-2,2,4-trimethyl-4-[(trimethylsilyl)oxy]-3,8-dioxa-2,4-disilaundecan-11-yl 2-methylprop-2-enoate ・Hexadecyl prop-2-enoate ・Docosyl prop-2-enoate ・1-ethylcyclohexyl 2-methylprop-2-enoate ・Docosyl prop-2-enoate - tetradecyl prop-2-enoate - adamantan-1-yl 2-methylprop-2-enoate - 3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate - 2-Propenoic acid isooctadecyl ester (ISTA) [Film-Like Resin] The film-like resin of this embodiment will be described with reference to FIG. 3. FIG. 3 is a schematic diagram showing an example of a film-like resin.

[0059] 3, the film-like resin in this embodiment is obtained by forming a photosensitive resin composition into a film shape. That is, a film-like resin 1 can be obtained by forming a photosensitive resin composition into a film shape.

[0060] [Resin Sheet] The resin sheet in this embodiment will be described with reference to Fig. 4. Fig. 4 is a schematic diagram showing an example of a resin sheet.

[0061] 4, the resin sheet 2 in this embodiment includes a substrate 3 and a resin layer 4. The resin layer 4 is formed into a film by applying a photosensitive resin composition onto the substrate 3 and drying it. That is, the resin layer 4 is made of a film-like resin 1 formed by forming the photosensitive resin composition into a film.

[0062] [Composition Proposal System] The composition of the base polymer contained in the photosensitive resin composition in this embodiment is proposed, for example, by a composition proposal system. The composition proposal system in this embodiment is an information processing system that proposes the composition of the base polymer contained in the photosensitive resin composition.

[0063] The composition suggestion system predicts the physical properties of the base polymer based on candidate compositions of the base polymer, and proposes candidate compositions that satisfy the target characteristics based on the predicted results. In this embodiment, the physical properties of the base polymer include physical properties related to resolution and physical properties related to adhesion. Therefore, the composition suggestion system can propose candidate compositions for the base polymer that achieve both high resolution and high adhesion.

[0064] <Overall Configuration> The overall configuration of the composition suggestion system in this embodiment will be described with reference to Fig. 5. Fig. 5 is a block diagram showing an example of the overall configuration of the composition suggestion system in this embodiment.

[0065] 5, the composition proposal system 1000 in this embodiment includes a proposal device 10 and a terminal device 20. The proposal device 10 and the terminal device 20 are connected to each other so as to be able to communicate data with each other via a communication network N1 such as a local area network (LAN) or the Internet.

[0066] The proposal device 10 is an information processing device such as a personal computer, workstation, or server that proposes a composition of a base polymer contained in a photosensitive resin composition. The proposal device 10 receives proposal conditions from the terminal device 20. The proposal conditions include production conditions for candidate compositions and target properties that the candidate compositions must satisfy. The proposal device 10 generates candidate compositions based on the production conditions and predicts the physical properties of the base polymer for each candidate composition. The proposal device 10 generates proposal information indicating candidate compositions that satisfy the target properties and transmits it to the terminal device 20.

[0067] The terminal device 20 is an information processing terminal such as a personal computer, smartphone, or tablet terminal operated by a user of the composition proposal system 1000. The terminal device 20 transmits proposed conditions indicating the generation conditions and target properties to the proposal device 10 in response to the user's operation. The terminal device 20 outputs the proposed information received from the proposal device 10 to the user.

[0068] A user of the composition proposal system 1000 can refer to the proposal information output to the terminal device 20 and create a base polymer according to the candidate composition indicated in the proposal information. The user can also manufacture a photosensitive resin composition using the base polymer created according to the candidate composition. The candidate composition indicated in the proposal information is a base polymer that realizes a photosensitive resin composition that achieves both high resolution and high adhesion, so the user can manufacture a photosensitive resin composition, a film-like resin, and a resin sheet that achieves both high resolution and high adhesion.

[0069] The overall configuration of the composition proposal system 1000 shown in Figure 5 is one example, and various system configuration examples are possible depending on the application and purpose. For example, the composition proposal system 1000 may include multiple proposal devices 10 and one or more terminal devices 20. For example, the proposal device 10 may be realized by multiple computers, or may be realized as a cloud computing service. The division of devices such as the proposal device 10 and terminal device 20 shown in Figure 5 is one example.

[0070] <Hardware Configuration> The hardware configuration of the composition suggestion system 1000 in this embodiment will be described with reference to FIG.

[0071] <Hardware Configuration of Computer> The proposal device 10 and the terminal device 20 in this embodiment are realized by, for example, a computer. Fig. 6 is a block diagram showing an example of the hardware configuration of a computer 500 in this embodiment.

[0072] 6 , the computer 500 includes a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, a HDD (Hard Disk Drive) 504, an input device 505, a display device 506, a communication I / F (Interface) 507, and an external I / F 508. The CPU 501, the ROM 502, and the RAM 503 form a so-called computer. The hardware components of the computer 500 are connected to each other via a bus line 509. The input device 505 and the display device 506 may be connected to the external I / F 508 for use.

[0073] The CPU 501 is a computing device that reads programs and data from a storage device such as the ROM 502 or the HDD 504 onto the RAM 503 and executes processing to realize overall control and functions of the computer 500 .

[0074] The ROM 502 is an example of a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The ROM 502 functions as a main storage device that stores various programs, data, etc. required for the CPU 501 to execute various programs installed in the HDD 504. Specifically, the ROM 502 stores boot programs such as a Basic Input / Output System (BIOS) and an Extensible Firmware Interface (EFI) that are executed when the computer 500 starts up, as well as data such as OS (Operating System) settings and network settings.

[0075] The RAM 503 is an example of a volatile semiconductor memory (storage device) in which programs and data are erased when the power is turned off. The RAM 503 is, for example, a dynamic random access memory (DRAM) or a static random access memory (SRAM). The RAM 503 provides a working area in which various programs installed in the HDD 504 are expanded when executed by the CPU 501.

[0076] The HDD 504 is an example of a non-volatile storage device that stores programs and data. The programs and data stored in the HDD 504 include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS. Note that the computer 500 may use a storage device that uses flash memory as a storage medium (e.g., an SSD (Solid State Drive)) instead of the HDD 504.

[0077] The input device 505 includes a touch panel, operation keys and buttons, a keyboard and mouse, a microphone for inputting sound data such as voice, and the like, which are used by the user to input various signals.

[0078] The display device 506 is composed of a display such as a liquid crystal display or organic electroluminescence (EL) display for displaying a screen, a speaker for outputting sound data such as voice, and the like.

[0079] The communication I / F 507 is an interface that connects to a communication network and enables the computer 500 to perform data communication.

[0080] The external I / F 508 is an interface with external devices, such as a drive device 510.

[0081] The drive device 510 is a device for loading a recording medium 511. The recording medium 511 here includes media that record information optically, electrically, or magnetically, such as CD-ROMs, flexible disks, and magneto-optical disks. The recording medium 511 may also include semiconductor memories that record information electrically, such as ROMs and flash memories. This allows the computer 500 to read from and / or write to the recording medium 511 via the external I / F 508.

[0082] The various programs to be installed in the HDD 504 are installed, for example, by setting the distributed recording medium 511 in a drive device 510 connected to the external I / F 508 and reading the various programs recorded on the recording medium 511 by the drive device 510. Alternatively, the various programs to be installed in the HDD 504 may be installed by being downloaded via the communication I / F 507 from a network different from the communication network.

[0083] <Functional Configuration> The functional configuration of the composition suggestion system in this embodiment will be described with reference to Fig. 7 to Fig. 9. Fig. 7 is a block diagram showing an example of the functional configuration of the composition suggestion system in this embodiment.

[0084] <Proposal Device> As shown in FIG. 7 , the proposal device 10 in this embodiment includes an experimental data storage unit 100, a descriptor calculation unit 101, a descriptor storage unit 102, a teacher data generation unit 103, a model learning unit 104, a model storage unit 105, a proposal condition acquisition unit 106, a candidate composition generation unit 107, an explanatory variable calculation unit 108, a physical property prediction unit 109, and a proposal information generation unit 110.

[0085] The experimental data storage unit 100, the descriptor storage unit 102, and the model storage unit 105 are realized by the HDD 504 shown in FIG.

[0086] The descriptor calculation unit 101, the teacher data generation unit 103, the model learning unit 104, the proposed condition acquisition unit 106, the candidate composition generation unit 107, the explanatory variable calculation unit 108, the physical property prediction unit 109, and the proposed information generation unit 110 are realized by processing that is executed by the CPU 501 using a program loaded from the HDD 504 shown in FIG. 6 onto the RAM 503.

[0087] Experimental data showing experimental results for various dry film resists is stored in the experimental data storage unit 100. The experimental data is data in which information showing the dry film resist used in the experiment is associated with the composition of the base polymer contained in the dry film resist (experimental composition) and physical property values ​​observed in the experiment (experimental values).

[0088] 8 is a diagram showing an example of experimental data in this embodiment. As shown in FIG. 8, the experimental data includes, as data items, information (film) indicating the dry film resist used in the experiment, the experimental composition, and experimental values. The experimental composition includes the blending amount of acrylic acid, the blending amount of styrene, the blending amount of a third component, and identification information identifying the monomer that is the third component. The experimental values ​​include 16 μm pitch adhesion and resolution 3x / x.

[0089] Examples of the identification information include a compound name, a structural formula, SMILES (Simplified Molecular Input Line Entry System) information, and ECFP (Extended Connectivity Circular Fingerprints) information. The identification information is not limited to these, and any information that can identify a monomer can be used.

[0090] Returning to FIG. 7 , the descriptor calculation unit 101 calculates descriptors for monomers that can constitute the base polymer. In this embodiment, the descriptor calculation unit 101 is provided with a list of monomers for which descriptors are to be calculated, which is stored in a storage unit such as the HDD 504. The list of monomers includes acrylic acid, styrene, and a third component. The descriptors for the monomers include HSP_δD, the number of carboxylic acids, logP, and polymer Tg for each monomer. The descriptor calculation unit 101 generates descriptor data in which descriptors for the monomers are associated with identification information that identifies the monomers.

[0091] The descriptor storage unit 102 stores the descriptor data generated by the descriptor calculation unit 101 .

[0092] 9 is a diagram showing an example of descriptor data in this embodiment. As shown in FIG. 9, the descriptor data has, as data items, identification information for identifying a monomer, HSP_δD, the number of carboxylic acids, logP, and polymer Tg. The descriptor data may also have data items for storing other descriptors.

[0093] 7, the training data generation unit 103 generates training data for training a prediction model. The prediction model is a machine learning model that predicts the physical property values ​​of the base polymer.

[0094] Specifically, the teacher data generation unit 103 calculates explanatory variables related to the experimental composition based on the experimental composition included in the experimental data read from the experimental data storage unit 100. The explanatory variables related to the experimental composition include HSP_δD, the number of carboxylic acids, log P, and polymer Tg. The teacher data generation unit 103 calculates HSP_δD, the number of carboxylic acids, log P, and polymer Tg related to the experimental composition by weighting and adding the HSP_δD, the number of carboxylic acids, log P, and polymer Tg of the monomers read from the descriptor storage unit 102 based on the blending amount of each monomer included in the experimental composition.

[0095] Next, the teacher data generation unit 103 acquires experimental values ​​included in the experimental data. Specifically, the teacher data generation unit 103 acquires the 16 μm pitch adhesion and resolution 3x / x included in the experimental data. The teacher data generation unit 103 then generates teacher data related to the experimental composition by combining explanatory variables related to the experimental composition with the experimental values ​​as correct values ​​of the objective variables.

[0096] The model learning unit 104 learns a prediction model based on the training data generated by the training data generation unit 103. The prediction model includes a prediction model (hereinafter referred to as a "resolution prediction model") that predicts a physical property value indicating resolution (i.e., resolution 3x / x) and a prediction model (hereinafter referred to as an "adhesion prediction model") that predicts a physical property value indicating adhesion (i.e., 16 μm pitch adhesion).

[0097] The model storage unit 105 stores the trained prediction model generated by the model training unit 104. That is, the model storage unit 105 stores a trained resolution prediction model and a trained adhesion prediction model.

[0098] The proposed condition acquisition unit 106 acquires proposed conditions indicating conditions related to the composition of the base polymer. The proposed condition acquisition unit 106 may acquire the proposed conditions by receiving information indicating the proposed conditions from the terminal device 20. The proposed condition acquisition unit 106 may acquire the proposed conditions by accepting the proposed conditions input from the input device 505. The proposed condition acquisition unit 106 may acquire the proposed conditions by reading out the proposed conditions stored in advance in a storage unit such as the HDD 504.

[0099] The proposed conditions include production conditions for the candidate composition and target properties that the candidate composition should satisfy. The production conditions include conditions regarding the blending amount of acrylic acid, conditions regarding the blending amount of styrene, and a list of monomers to be used as the third component. The target properties include target values ​​for physical properties related to resolution and target values ​​for physical properties indicating adhesion. All or part of the proposed conditions may be determined in advance and stored in a memory unit.

[0100] The candidate composition generating unit 107 generates a candidate composition based on the production conditions included in the proposed conditions acquired by the proposed condition acquiring unit 106. Specifically, the candidate composition generating unit 107 generates options for the blending amount of acrylic acid and options for the blending amount of styrene in accordance with the production conditions. Next, the candidate composition generating unit 107 combines the blending amount of acrylic acid with the blending amount of styrene to calculate the blending amount of the third component. Then, the candidate composition generating unit 107 combines all of the monomers included in the list as the third component with each combination of blending amounts. In this way, all of the candidate compositions are generated.

[0101] The explanatory variable calculation unit 108 calculates explanatory variables related to the candidate composition based on the candidate composition generated by the candidate composition generation unit 107. The method by which the explanatory variable calculation unit 108 calculates the explanatory variables is the same as the method by which the training data generation unit 103 calculates the explanatory variables.

[0102] The physical property value prediction unit 109 predicts physical property values ​​related to the candidate composition by inputting the explanatory variables calculated by the explanatory variable calculation unit 108 into a prediction model read out from the model storage unit 105. Specifically, the physical property value prediction unit 109 predicts physical property values ​​related to resolution by inputting the explanatory variables related to the candidate composition into a trained resolution prediction model. Next, the physical property value prediction unit 109 predicts physical property values ​​related to adhesion by inputting the explanatory variables related to the candidate composition into a trained adhesion prediction model.

[0103] The proposal information generating unit 110 generates proposal information based on the physical property values ​​predicted by the physical property value predicting unit 109. The proposal information generating unit 110 transmits the proposal information to the terminal device 20.

[0104] The proposed information includes information on candidate compositions whose physical property values ​​predicted by the physical property value prediction unit 109 satisfy the target characteristics included in the proposed conditions. The proposed information may also include information on candidate compositions whose physical property values ​​predicted by the physical property value prediction unit 109 do not satisfy the target characteristics included in the proposed conditions.

[0105] For example, the proposal information generating unit 110 may extract candidate compositions whose physical property values ​​satisfy the target characteristics, and generate proposal information in which the extracted candidate compositions are arranged according to a predetermined rule (for example, ascending or descending order of physical property values, ascending or descending order of the blending amount of a specific monomer, etc.). Furthermore, for example, the proposal information generating unit 110 may generate proposal information in which all candidate compositions are arranged in order of favorable physical property values.

[0106] <<Terminal Device>> As shown in FIG. 7 , the terminal device 20 in this embodiment includes a proposal condition input unit 201 and a proposal information display unit 202 .

[0107] The proposal condition input unit 201 and the proposal information display unit 202 are realized by the processing that the CPU 501 executes by a program loaded from the HDD 504 onto the RAM 503 shown in FIG.

[0108] The proposal condition input unit 201 receives input of proposal conditions in response to a user operation, and transmits information indicating the received proposal conditions to the proposal device 10.

[0109] The proposal information display unit 202 receives proposal information from the proposal device 10. The proposal information display unit 202 outputs the received proposal information to the display device 506.

[0110] <Prediction Model> The prediction model in this embodiment will be described in more detail with reference to FIGS.

[0111] 10 is a block diagram showing an example of an adhesion prediction model 301. As shown in FIG. 10, the adhesion prediction model 301 is a machine learning model that uses the number of carboxylic acids, HSP_δD, and logP as explanatory variables and 16 μm pitch adhesion as a response variable.

[0112] The adhesion prediction model 301 is, for example, a machine learning model based on Gaussian process regression. When implemented with scikit-learn, the kernel function can be, for example, ConstantKernel() * RBF(np.ones(3)) + WhiteKernel(). Note that scikit-learn is an open-source machine learning library.

[0113] FIG. 11 is a graph showing an example of the prediction accuracy of the adhesion prediction model. FIG. 11 shows the results of evaluating the prediction accuracy of the trained adhesion prediction model 301 by cross-validation. Here, the coefficient of determination (R2) and root mean squared error (RMSE) were used as evaluation indices. As shown in FIG. 11, R2 was 0.95 and RMSE was 0.72. FIG. 11 shows that the adhesion prediction model 301 can predict 16 μm pitch adhesion with high accuracy.

[0114] 12 is a block diagram showing an example of a resolution prediction model 302. As shown in FIG. 12, the resolution prediction model 302 is a machine learning model that uses the number of carboxylic acids, HSP_δD, and polymer Tg as explanatory variables and the resolution 3x / x as a response variable.

[0115] The resolution prediction model 302 is, for example, a machine learning model based on Gaussian process regression. When implemented using scikit-learn, the kernel function may be, for example, ConstantKernel() * RBF() + WhiteKernel().

[0116] Fig. 13 is a graph showing an example of the prediction accuracy of the resolution prediction model. Fig. 13 shows the results of evaluating the prediction accuracy of the trained resolution prediction model 302 by cross-validation. Here, as in Fig. 11 , R2 and RMSE were used as evaluation indices. As shown in Fig. 13, R2 was 0.83 and RMSE was 1.03. Fig. 13 shows that the resolution prediction model 302 can predict the resolution 3x / x with high accuracy.

[0117] <Processing Procedure> The processing procedure of the proposal method executed by the composition proposal system 1000 in this embodiment will be described with reference to Fig. 14 to Fig. 16. The proposal method in this embodiment includes a descriptor calculation process (see Fig. 14), a model learning process (see Fig. 15), and a composition proposal process (see Fig. 16).

[0118] <Descriptor Calculation Process> Fig. 14 is a flowchart showing an example of the descriptor calculation process in this embodiment. The descriptor calculation process is a process for generating descriptor data related to a monomer.

[0119] In step S1, the descriptor calculation unit 101 of the proposal device 10 reads out a list of monomers from the storage unit 10. Next, the descriptor calculation unit 101 acquires a monomer for which a descriptor is to be calculated from the read out list.

[0120] In step S2, the descriptor calculation unit 101 of the proposed device 10 calculates a descriptor for the monomer acquired in step S1. Specifically, the descriptor calculation unit 101 calculates the HSP_δD, the number of carboxylic acids, logP, and polymer Tg for the monomer.

[0121] In step S3, the descriptor calculation unit 101 of the proposed device 10 associates the descriptor calculated in step S2 with identification information that identifies the monomer. This generates descriptor data. Next, the descriptor calculation unit 101 stores the descriptor data in the descriptor storage unit 102.

[0122] The processes from step S2 to step S3 are repeatedly executed for all polymers included in the list read out in step S1.

[0123] <Model Learning Process> Fig. 15 is a flowchart showing an example of the model learning process according to this embodiment. The model learning process is a process for learning a prediction model.

[0124] In step S11, the teacher data generating unit 103 of the proposal device 10 reads out the experimental data from the experimental data storage unit 100. Next, the teacher data generating unit 103 selects an experimental composition to be used for learning from the experimental data.

[0125] In step S12, the teacher data generation unit 103 of the proposal device 10 reads out descriptor data for each monomer included in the experimental composition selected in step S11 from the descriptor storage unit 102. Next, the teacher data generation unit 103 obtains HSP_δD, the number of carboxylic acids, logP, and polymer Tg from the read-out descriptor data.

[0126] In step S13, the teacher data generation unit 103 of the proposal device 10 calculates explanatory variables related to the experimental composition based on the descriptors acquired in step S12. Specifically, the teacher data generation unit 103 weights and adds the HSP_δD, number of carboxylic acids, log P, and polymer Tg related to the monomers based on the blending amount of each monomer contained in the experimental composition. This allows the HSP_δD, number of carboxylic acids, log P, and polymer Tg related to the experimental composition to be calculated.

[0127] Then, the teacher data generation unit 103 acquires experimental values ​​corresponding to the experimental composition selected in step S11. Specifically, the teacher data generation unit 103 acquires the 16 μm pitch adhesion and resolution 3x / x included in the experimental data. Next, the teacher data generation unit 103 combines the experimental values ​​as the correct values ​​of the objective variables with the explanatory variables related to the experimental composition. This generates teacher data related to the experimental composition. The teacher data generation unit 103 sends the teacher data related to the experimental composition to the model learning unit 104.

[0128] In step S14, the model learning unit 104 of the proposal device 10 receives training data related to the experimental composition from the training data generation unit 103. Next, the model learning unit 104 predicts physical property values ​​related to the experimental composition by inputting explanatory variables included in the training data into the prediction model being trained.

[0129] Specifically, the model learning unit 104 first inputs the number of carboxylic acids, HSP_δD, and logP contained in the training data into the adhesion prediction model under training. The adhesion prediction model predicts 16 μm pitch adhesion based on the input number of carboxylic acids, HSP_δD, and logP, and outputs the predicted value.

[0130] Next, the model learning unit 104 inputs the number of carboxylic acids, HSP_δD, and polymer Tg included in the training data into the resolution prediction model being trained. The resolution prediction model predicts the resolution 3x / x based on the input number of carboxylic acids, HSP_δD, and polymer Tg, and outputs the predicted value.

[0131] In step S15, the model learning unit 104 of the proposal device 10 calculates the error between the predicted value predicted in step S14 and the correct value included in the training data. Next, the model learning unit 104 updates the parameters of the prediction model so as to minimize the error between the predicted value and the correct value.

[0132] Specifically, the model learning unit 104 calculates the error between the predicted value of 16 μm pitch adhesion output from the adhesion prediction model in step S14 and the correct value of 16 μm pitch adhesion included in the training data, and updates the parameters of the adhesion prediction model so as to minimize the error. Also, the model learning unit 104 calculates the error between the predicted value of resolution 3x / x output from the resolution prediction model in step S14 and the correct value of resolution 3x / x included in the training data, and updates the parameters of the resolution prediction model so as to minimize the error.

[0133] In step S16, the model learning unit 104 of the proposal device 10 determines whether learning is complete. Whether learning is complete may be determined based on whether the error calculated in step S14 has converged, or whether a predetermined number of repetitions has been exceeded. If learning is not complete (NO), the model learning unit 104 returns the process to step S15. If learning is not complete (NO), the model learning unit 104 may return the process to step S11 to acquire more experimental data. In this case, the teacher data generation unit 103 selects a different experimental composition and executes the processes from steps S15 to S16 again. If learning is complete (YES), the model learning unit 104 ends the model learning process.

[0134] <<Composition Proposal Process>> Fig. 16 is a flowchart showing an example of the composition proposal process in this embodiment. The composition proposal process is a process for proposing candidate compositions that satisfy target properties based on a trained prediction model.

[0135] In step S21, the proposal condition input unit 201 of the terminal device 20 accepts input of proposal conditions in response to a user operation. The proposal conditions include conditions for generating a candidate composition and target properties that the candidate composition should satisfy. All or part of the proposal conditions may be determined in advance. If all of the proposal conditions are determined in advance, the processing of step S21 does not need to be executed.

[0136] When the proposal condition input unit 201 accepts the proposal conditions, it transmits information indicating the proposal conditions to the proposal device 10. In the proposal device 10, the proposal condition acquisition unit 106 receives the information indicating the proposal conditions from the terminal device 20.

[0137] In step S22, the proposed condition acquisition unit 106 of the proposing device 10 acquires the production conditions included in the information indicating the proposed conditions and sends them to the candidate composition generation unit 107. The proposed condition acquisition unit 106 also acquires the target properties included in the information indicating the proposed conditions and sends them to the proposed information generation unit 110.

[0138] In step S23, the candidate composition generator 107 of the proposing device 10 receives the composition conditions from the proposed condition acquisition unit 106. Next, the candidate composition generator 107 generates a candidate composition based on the composition conditions. The candidate composition generator 107 sends the generated candidate composition to the explanatory variable calculation unit 108.

[0139] In step S24, the explanatory variable calculation unit 108 of the proposing device 10 receives the candidate composition from the candidate composition generation unit 107. Next, the explanatory variable calculation unit 108 calculates explanatory variables related to the candidate composition based on the candidate composition. The method for calculating the explanatory variables is the same as the method used by the training data generation unit 103 to calculate the explanatory variables in step S13. The explanatory variable calculation unit 108 sends the explanatory variables related to the candidate composition to the physical property value prediction unit 109.

[0140] In step S25, the physical property prediction unit 109 of the proposing device 10 receives explanatory variables related to the candidate composition from the explanatory variable calculation unit 108. Next, the physical property prediction unit 109 reads out the trained adhesion prediction model and the trained resolution prediction model from the model storage unit 105.

[0141] Next, the physical property prediction unit 109 inputs the number of carboxylic acids, HSP_δD, and logP for the candidate composition into the trained adhesion prediction model. The adhesion prediction model predicts 16 μm pitch adhesion based on the input number of carboxylic acids, HSP_δD, and logP, and outputs the predicted value.

[0142] The physical property prediction unit 109 also inputs the number of carboxylic acids, HSP_δD, and polymer Tg for the candidate composition into the trained resolution prediction model. The resolution prediction model predicts the resolution 3x / x based on the input number of carboxylic acids, HSP_δD, and polymer Tg, and outputs the predicted value.

[0143] Then, the physical property value prediction unit 109 sends the physical property values ​​related to the candidate composition to the proposal information generation unit 110. Specifically, the physical property value prediction unit 109 sends the predicted value of 16 μm pitch adhesion output by the adhesion prediction model and the predicted value of resolution 3x / x output by the resolution prediction model to the proposal information generation unit 110.

[0144] In step S26, the proposal information generating unit 110 of the proposing device 10 receives the target properties from the proposal condition acquiring unit 106. The proposal information generating unit 110 also receives the physical property values ​​related to the candidate composition from the physical property predicting unit 109.

[0145] Next, the proposal information generator 110 determines whether the physical property values ​​of the candidate composition satisfy the target characteristics. Specifically, the proposal information generator 110 determines whether the predicted value for 16 μm pitch tightness is less than the target value for 16 μm pitch tightness included in the target characteristics. The proposal information generator 110 also determines whether the predicted value for resolution 3x / x is less than the target value for resolution 3x / x included in the target characteristics.

[0146] The proposal information generator 110 determines that the physical property values ​​of the candidate composition satisfy the target characteristics when both the predicted values ​​for 16 μm pitch adhesion and the predicted values ​​for resolution 3x / x are less than the target values. On the other hand, the proposal information generator 110 determines that the physical property values ​​of the candidate composition do not satisfy the target characteristics when at least one of the predicted values ​​for 16 μm pitch adhesion and the predicted values ​​for resolution 3x / x is equal to or greater than the target value.

[0147] If the proposal information generator 110 determines that the physical property values ​​of the candidate composition satisfy the target characteristics (YES), the proposal information generator 110 proceeds to step S27. On the other hand, if the proposal information generator 110 determines that the physical property values ​​of the candidate composition do not satisfy the target characteristics (NO), the proposal information generator 110 skips step S27.

[0148] If the proposed information includes a candidate composition that does not satisfy the target properties, step S26 does not need to be performed.

[0149] In step S27, the proposal information generator 110 of the proposing device 10 adds information about the candidate composition whose physical property values ​​are determined to satisfy the target characteristics to the proposal information. The proposal information includes the candidate composition and the physical property values. The candidate composition includes the blending amount of acrylic acid, the blending amount of styrene, the blending amount of a third component, and identification information for identifying the monomer that is the third component. The physical property values ​​include predicted values ​​for 16 μm pitch adhesion and predicted values ​​for resolution 3x / x.

[0150] The processes from step S24 to step S27 are repeatedly executed for all the candidate compositions generated in step S23.

[0151] In step S28, the proposal information generation unit 110 of the proposal device 10 transmits the proposal information to the terminal device 20. In the terminal device 20, the proposal information display unit 202 receives the proposal information from the proposal device 10.

[0152] In step S29, the proposed information display unit 202 of the terminal device 20 displays the proposed information on the display device 506. The user can refer to the proposed information displayed on the display device 506 of the terminal device 20 and produce a base polymer to be contained in the photosensitive resin composition according to the candidate composition indicated in the proposed information. The user can search for a more appropriate base polymer composition by examining the proposed information displayed on the display device 506 of the terminal device 20 and inputting new proposed conditions.

[0153] Effect of the embodiment The photosensitive resin composition of the present embodiment contains a polymer composed of multiple types of monomers, and predetermined explanatory variables calculated based on the monomers fall within a predetermined range. The predetermined range is determined based on a composition in which the predicted results of a machine learning model using physical property values ​​related to resolution and physical property values ​​related to adhesion as objective variables satisfy predetermined target values. Therefore, according to the present embodiment, a photosensitive resin composition, a film-like resin, and a resin sheet having excellent resolution and adhesion are provided.

[0154] The proposal device 10 in this embodiment predicts physical property values ​​related to resolution and adhesion for candidate compositions generated under predetermined production conditions based on a trained prediction model, and proposes candidate compositions whose prediction results satisfy predetermined target characteristics. Therefore, according to this embodiment, it is possible to propose a base polymer composition that realizes a photosensitive resin composition with excellent resolution and adhesion.

[0155] Although the embodiments of the present disclosure have been described in detail above, the embodiments disclosed herein are illustrative in all respects and are not limiting. The embodiments can be modified and improved in various ways without departing from the scope and spirit of the appended claims. The matters described in the above embodiments can be configured in other ways as long as they are not inconsistent, and can be combined as long as they are not inconsistent.

[0156] This application claims priority from Japanese Patent Application No. 2023-104223, filed on June 26, 2023, with the Japan Patent Office, the entire contents of which are incorporated herein by reference.

[0157] REFERENCE SIGNS LIST 1 Film-like resin 2 Resin sheet 3 Base material 4 Resin layer 10 Proposal device 100 Experimental data storage unit 101 Descriptor calculation unit 102 Descriptor storage unit 103 Teacher data generation unit 104 Model learning unit 105 Model storage unit 106 Proposal condition acquisition unit 107 Candidate composition generation unit 108 Explanatory variable calculation unit 109 Physical property value prediction unit 110 Proposal information generation unit 20 Terminal device 201 Proposal condition input unit 202 Proposal information display unit 1000 Composition proposal system

Claims

1. A photosensitive resin composition containing a polymer composed of multiple types of monomers, wherein the variance term of the Hansen solubility parameter calculated based on the monomers is 16.99 or more and 17.35 or less, the number of carboxylic acids in the molecule calculated based on the monomers is 0.272 or more and 0.304 or less, the logarithm of the octanol / water partition coefficient calculated based on the monomers is 2.16 or more and 2.79 or less, and the glass transition temperature calculated based on the monomers is 354 K or more and 382 K or less.

2. The photosensitive resin composition according to claim 1, wherein the dispersion term of the Hansen solubility parameter is 17.04 or more and 17.19 or less, the number of carboxylic acids is 0.273 or more and 0.290 or less, the logarithm of the octanol / water partition coefficient is 2.29 or more and 2.78 or less, and the glass transition temperature is 365 K or more and 382 K or less.

3. The photosensitive resin composition according to claim 1 or 2, wherein the monomer comprises acrylic acid and styrene.

4. The photosensitive resin composition according to claim 1, wherein the monomer contains a third component different from both acrylic acid and styrene, and the third component is 3-phenoxybenzyl acrylate, 3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate, octadecyl 2-methylprop-2-enoate, 4-hydroxyphenyl 2-methyl-2-propenoate, octadecyl 2-methylprop-2-enoate, dicyclopentanyl methacrylate, decyl prop-2-enoate, dodecyl acrylate, or isooctadecyl 2-propenoate.

5. The monomer includes a third component different from both acrylic acid and styrene, and the third component is 10-hydroxy-2,2,4-trimethyl-4-[(trimethylsilyl)oxy]-3,8-dioxa-2,4-disilaundecan-11-yl 2-methylprop-2-enoic acid, hexadecyl prop-2-enoate, docosylprop-2-enoic acid, 1-ethylcyclohexyl 2-methylprop-2-enoate, docosylprop-2-enoic acid, tetradecyl prop-2-enoate, adamantan-1-yl 2-methylprop-2-enoate, 3-{2,2,6,6-tetramethyl-4-[(trimethylsilyl)oxy]-3,5-dioxa-2,4,6-trisilaheptan-4-yl}propyl 2-methylprop-2-enoate, or isooctadecyl 2-propenoate. The photosensitive resin composition according to claim 2.

6. The photosensitive resin composition according to claim 4 or 5, wherein the polymer contains the third component in an amount of 13% by weight or more and 22% by weight or less.

7. A film-like resin obtained by forming the photosensitive resin composition according to any one of claims 1 to 6 on a film.

8. A resin sheet comprising a substrate and a resin layer formed on the substrate, the resin layer being made of the photosensitive resin composition according to any one of claims 1 to 6.

Images