System for predicting adhesion state, method for predicting adhesion state, program for predicting adhesion state, and method for producing adherend

Abstract

The invention relates to an adhesion state prediction system, an adhesion state prediction method, an adhesion state prediction program, and a method for manufacturing an adherend. This adhesion state prediction system predicts the adhesion state when an adhesive object is adhered to an adherend, and is provided with: a data acquisition unit that acquires two or more pieces of data including physical property value information on two-dimensional coordinates of the object; and an adhesion state prediction unit that predicts the adhesion state between the object and the adherend using the two or more acquired data.

Description

Technical Field

[0001] The present invention relates to a system for predicting an adhesion state, a method for predicting an adhesion state, a program for predicting an adhesion state, and a method for manufacturing an adhered product. Background Art

[0002] In recent years, in the industrial world, not only creating products and services that meet customer needs, but also achieving SDGs (Sustainable Development Goals) has become increasingly important as a driving force for corporate activities.

[0003] For example, in SDGs 9-4, "Increase sustainability through infrastructure improvement and industrial improvement through the introduction of clean technology and environmentally friendly technology and industrial processes through the improvement of resource utilization efficiency." This will increase the need for technologies to reduce losses in the manufacturing process, improve product yields, and detect product defects in upstream processes.

[0004] Such SDGs are expected to be achieved through DX (digital transformation) and Society 5.0 (a new super-intelligent society brought about by the interconnected industries where various objects and things are connected). Therefore, algorithms that can judge and select quickly and efficiently are required to be provided as data in a processable form and applied to the process.

[0005] In particular, the bonding process is cited as one of the reasons for the reduction in the yield of various industrial products. This is because, generally speaking, defective products produced in the bonding process cannot be returned to the original parts and products and need to be discarded. For example, in the mobile body industry such as automobiles and aircraft, lightweighting with the purpose of reducing the use of fossil fuels requires a shift from metal to carbon reinforced fiber and from metal bolts to adhesives. In addition, in lithium-ion batteries, which are the key to electric vehicles with low environmental load, higher durability, that is, high-precision bonding force prediction, is also required from the perspective of ensuring safety. In addition, in the electronics industry and display industry where various new materials and high-priced components are introduced, the product development cycle is short and is used by a large number of people, so the importance of the subject increases.

[0006] For example, in recent years, as a change point in the display industry, with the change of form factors such as foldable, new problems such as degradation and peeling when folded have arisen. Therefore, more efficient development methods are needed than before. These electronic devices usually have a stacked structure composed of multiple layers, and the bonding technology between adjacent layers is very important.

[0007] At present, in order to determine the bonding state, it is necessary to extract samples from the manufacturing process and conduct destructive tests. In the case of trial production in a test factory for the purpose of formula optimization, problem proposal, etc., in order to confirm the performance, the only technology used in the past was to inspect the final sample through a destructive test. In this way, after analyzing the inspection results of the final sample, feedback is given, and trial production is carried out again. This cycle has poor development efficiency and cannot keep up with the speed of product development in the market. On the other hand, in the trial production process, if the problem is extracted in real time, for example, the optimal formula can be obtained quickly, thus bringing about efficiency in development.

[0008] In contrast, as an inspection system, a system is disclosed that obtains three-dimensional information formed by combining position information in a two-dimensional space with spectral data corresponding to each position to visualize the inspection object and evaluate the adhesion amount / film thickness (for example, Patent Document 1). However, this system only visualizes the state of the inspection object and cannot predict defects (for example, poor adhesion, poor curing) that may occur in the inspection object based on the acquired information.

[0009] On the other hand, as a method for predicting the occurrence of poor adhesion in advance, a method for predicting poor curing of a deteriorated portion by observing the optical properties of a bonding layer composed of a resin composition in an uncured state is disclosed, wherein the resin composition includes a synthetic resin cured by ring-opening polymerization and a pH indicator whose optical properties change according to pH (for example, Patent Document 2).

[0010] In addition, a method for estimating a curing state is disclosed, which includes: a step of irradiating ultraviolet rays to an ultraviolet curing resin containing a main agent and a photopolymerization initiator; a step of detecting fluorescence released by the photopolymerization initiator when receiving the ultraviolet rays; and a step of estimating the curing state of the ultraviolet curing resin based on the detected fluorescence (for example, patent document 3).

[0011] Patent Document 1: Japanese Patent Application Publication No. 2019-191130

[0012] Patent Document 2: Japanese Patent Application Publication No. 2020-94083

[0013] Patent Document 3: Japanese Patent Application Publication No. 2007-248244

[0014] However, in most cases, the occurrence of abnormal bonding state involves various factors. Therefore, it is difficult to predict the occurrence of abnormal bonding state by capturing only one phenomenon. Therefore, it is desirable to capture two or more phenomena or the temporal changes of a specific phenomenon, that is, to obtain multivariate data.

[0015] Furthermore, after acquiring the multidimensional data as described above, it is desirable to further analyze and predict the bonding state using an algorithm suitable for each material.

[0016] In contrast, the method of Patent Document 2 is only a limited technology that can be applied to a specific material (synthetic resin cured by ring-opening polymerization) only for a specific reason (salt adhesion). In addition, the data obtained are only data about fluorescence, not multivariate data. The data related to fluorescence intensity and its time variation obtained in Patent Document 3 are local data, not data containing position information in two-dimensional space. Therefore, the occurrence of abnormal bonding state cannot be predicted with high accuracy. As such, the current situation is that there is no known method or system for predicting bonding state that can be applied to various bonding materials and with high accuracy. Summary of the invention

[0017] The present invention is accomplished in view of these circumstances, and its purpose is to provide a prediction system for the bonding state that can be applied to various bonding materials and can predict the bonding state with high accuracy. In addition, the present invention also aims to provide a prediction method for the bonding state, a prediction program for the bonding state, and a method for manufacturing an adhesive.

[0018] The bonding state prediction system of the present invention is a system for predicting the bonding state when an adhesive object is bonded to an adherend, wherein it comprises: a data acquisition unit for acquiring two or more data containing physical property value information on the two-dimensional coordinates of the above-mentioned object; and a bonding state prediction unit for predicting the bonding state between the above-mentioned object and the above-mentioned adherend using the two or more acquired data.

[0019] The method for predicting the bonding state of the present invention is a method for predicting the bonding state when an adhesive object is bonded to an adherend, which comprises: a process of acquiring two or more data containing physical property value information on the two-dimensional coordinates of the above-mentioned object; and a process of predicting the bonding state between the above-mentioned object and the above-mentioned adherend using the two or more acquired data.

[0020] The bonding state prediction program of the present invention is a program for predicting the bonding state when an adhesive object is bonded to an adherend, and is used to enable a computer to execute: a process of acquiring two or more data containing physical property value information on the two-dimensional coordinates of the above-mentioned object; and a process of predicting the bonding state between the above-mentioned object and the above-mentioned adherend using the two or more acquired data.

[0021] The method for producing an adhesive article of the present invention comprises: a step of predicting the adhesive state between the object and an adherend by performing the prediction method of the present invention on an adhesive object; and a step of adjusting the processing conditions of the object based on the prediction result.

[0022] According to the present invention, it is possible to provide a bonding state prediction system, a prediction method, a prediction program, and a method for producing a bonded product using the same, which are applicable to various bonding materials and can predict the bonding state with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flowchart showing an example of a method for predicting an adhesion state according to one embodiment of the present invention.

[0024] Figure 2A The image is shown when the temperature of the object containing the contrast medium (1) is changed, and the image is captured by a conventional camera at each temperature. Figure 2B The spectral spectrum obtained by averaging the spectral spectrum of a specific area captured by the hyperspectral camera at each temperature is shown.

[0025] Figure 3 It is a graph showing the behavior of the ratio of the emission intensity at a wavelength of 488 nm to the emission intensity at a wavelength of 590 nm (F488 / F590) obtained from the spectral spectrum when the temperature of the object is changed, and the behavior of the elastic modulus (Log value) measured separately.

[0026] Figure 4 This is a flowchart showing an example of the prediction process.

[0027] Figure 5 This is a flowchart showing a method for predicting an adhesion state according to another embodiment of the present invention.

[0028] Figure 6 This is a schematic diagram showing an example of the configuration of a system for predicting an adhesion state according to one embodiment of the present invention.

[0029] Figure 7 This is a flowchart showing an example of a method for producing a bonded product using the method for predicting a bonded state according to one embodiment of the present invention. DETAILED DESCRIPTION

[0030] Hereinafter, the present invention will be described in detail based on the embodiments, but the present invention is not limited to these embodiments.

[0031] 1. Prediction method of bonding state

[0032] First, a method for predicting an adhesion state according to an embodiment of the present invention will be described, and then a prediction system that can be used for the prediction method will be described.

[0033] A method for predicting an adhesive state according to one embodiment of the present invention is a method for predicting the adhesive state when an adhesive object and an adherend are bonded together, for example, the quality of the final adhesive state.

[0034] The object with adhesiveness is not particularly limited as long as it is an object showing adhesiveness, and can be any one of a thermocompression bonding material, an adhesive material, and a curable material. The thermocompression bonding material is a material that melts and adheres by heat, and examples thereof include low-density polyethylene, ethylene-vinyl acetate copolymer, polypropylene, etc. As adhesive materials, for example, acrylic, silicon, polyurethane, rubber and other adhesives are listed. As curable materials, photocurable materials and thermosetting materials are listed. The material constituting the adherend is also not particularly limited, and can be any one of glass, resin materials, and metal materials.

[0035] Figure 1 FIG. 4 is a flowchart showing an example of a method for predicting an adhesion state according to an embodiment of the present invention. Figure 1 As shown, the method for predicting the bonding state of the present embodiment includes: a process of acquiring two or more data containing physical property value information on the two-dimensional coordinates of the object (data acquisition process, steps S11 to S14); and a process of predicting the bonding state between the object and the adherend using the two or more acquired data (prediction process, step S15).

[0036] (Data acquisition process)

[0037] In the data acquisition step, two or more data including physical property value information on two-dimensional coordinates of the object are acquired.

[0038] "Data containing physical property value information on two-dimensional coordinates" (hereinafter also referred to as "data" or "data containing physical property value information") is data containing position information on two-dimensional coordinates and physical property value information corresponding to each position, that is, two-dimensional data of physical property value information.

[0039] Physical property value information refers to the physical property value of an object or information related thereto. The type of physical property value information is not particularly limited as long as it is a type related to the prediction of the bonding state, for example, including elastic modulus, degree of cure, hardness, film thickness, water content, residual solvent content, uneven coating, temperature, viscosity, dynamic viscoelasticity (storage modulus, loss elastic modulus), tanδ, surface tension, density, vapor pressure, boiling point, refractive index, cure shrinkage, glass transition temperature (Tg), SP value (polar component (dP), variance component (dD), hydrogen bond component (dH)), molecular weight (number average molecular weight Mn, weight average molecular weight Mw, polydispersity Mw / Mn), molecular structure information (functional group, chemical bonding state, free radical generation state), etc. Among them, from the perspective of predicting the bonding state with higher accuracy, it is preferred that the physical property value information is elastic modulus, degree of cure, hardness, uneven coating, polarity, water content, temperature or film thickness. Most of these physical property value information can be associated with the spectral characteristic information obtained from the spectral spectrum data. Therefore, it is preferable that at least one of the two or more data includes spectral characteristic information.

[0040] It is possible to obtain the two-dimensional coordinate information of the object, and to obtain data containing such physical property value information in correspondence with the obtained two-dimensional coordinate information. The two-dimensional coordinate information and the physical property value information can be obtained separately or simultaneously. In the case of simultaneous acquisition, it can be obtained directly or indirectly from the two-dimensional image. Examples of two-dimensional images include spectral images (multi-spectral images, hyperspectral images), reflectivity distribution images, temperature distribution images, etc.

[0041] The two or more data may be obtained by acquiring data containing specified physical property value information over time, or by acquiring two or more different physical property value information. The data acquired over time may be data acquired intermittently or continuously. The data acquired by acquiring two or more different physical property value information may be acquired simultaneously or at different timings.

[0042] At least one of the two or more data preferably includes spectral characteristic information of the object as described above. The spectral characteristic information may be the luminous intensity or luminous intensity ratio, peak shift, reflectivity, etc. at a specific wavelength obtained from the spectral spectrum data. Preferably, the spectral characteristic information is associated with one or more selected from the group consisting of elastic modulus, degree of curing, hardness, polarity, and water content.

[0043] Data containing such spectral characteristic information can be obtained based on the state of light reflected / released from the object when irradiating light of a specified wavelength to the object. Data containing spectral characteristic information can also be obtained separately by linking two-dimensional coordinate information and the corresponding information of light reflected / released from the object as described above; or it can be obtained simultaneously as a spectral image. The spectral image can be obtained by a device capable of detecting light reflected / released from the object, such as a hyperspectral camera. In this case, the object is preferably an object whose luminous behavior changes depending on its state. "The state of the object" refers to one of the physical property value information of the object. In addition, "luminous behavior change" refers to any one or more changes in the peak wavelength, intensity, spectrum, fluorescence lifetime, phosphorescence lifetime, etc. of the light reflected / released by the object. The light emitted by the object can be the light emitted by the object itself, or it can be fluorescence or phosphorescence generated by the excitation of the luminescent substance contained in the object.

[0044] Here, an absorbing / luminescent substance contained in an object and whose absorbing / luminescent behavior (wavelength and intensity) changes depending on the state of the object is particularly referred to as a "contrast agent".

[0045] That is, the object preferably contains a contrast agent. The contrast agent may be a substance originally contained in the object, or a substance artificially added later. In the case where the contrast agent is a fluorescent / phosphorescent substance, in order to emit light, it is necessary to excite it by a known means, which means are not limited, and light excitation, current excitation, chemical excitation, thermal excitation, etc. can be used. Preferably, the contrast agent is a material that is excited and emits light by irradiating light of a specified wavelength, and more preferably, a material that is excited and emits fluorescence by irradiating light of a specified wavelength.

[0046] As a contrast agent whose luminescent behavior changes according to the state of the object, a known color-changing pigment can be generally used as long as it is a color-changing pigment that responds to the state to be observed. As color-changing pigments, there are photochromic pigments described in Adv. Mater., 2013, 25, p378, Japanese Patent Publication No. 2012-172139, Japanese Patent Publication No. 2019-38973, etc., lyotropic pigments described in Acc. Chem. Res., 2017, 50, p366, Japanese Patent Publication No. 2008-291210, WO2020 / 171199, etc., Japanese Patent Publication No. 2019-31606, Japanese Patent Publication No. 2015-53 3892, etc., the thermochromic dyes described in Chem. Soc. Rev., 1997, 26, p147, WO2008 / 007563, JP-A-2011-227462, etc., the piezochromic dyes described in Chem. Eur. J., 2012, 18, p4558, JP-A-2014-517711, Chem. Sci., 2020, 11, p7525, etc., but are not limited to these.

[0047] The contrast agent may be added in an amount that does not affect adhesion and other required performance and that enables detection of the adhesion state, and preferably is 10 ppm to 1.0% by weight, more preferably 50 ppm to 0.5%, and even more preferably 100 ppm to 0.1%.

[0048] As contrast agents, compounds whose luminescence behavior changes depending on the hardness of an object and compounds whose luminescence behavior changes depending on the water content, polarity, etc. in an object are exemplified for explanation.

[0049] As a compound whose luminescence behavior changes depending on the hardness of the object, for example, a phenazine compound such as a contrast agent (1) is listed. It is known that when the compound becomes excited by light, it passes through two or more excited states and emits light at different wavelengths, depending on whether the surrounding environment of the compound itself is "an environment that is easy to relax the structure". The "environment that is easy to relax the structure" mentioned here refers to an environment that comprehensively reflects any one or more of the following: a free volume around the contrast agent itself is sufficient and easy to move, heat energy is obtained that facilitates molecular movement, the surrounding viscosity is low, and the surrounding is not solid but liquid.

[0050] The state of small free volume and high viscosity, in other words, high elastic modulus and high degree of curing in terms of resin. That is, by observing the emission wavelength of the contrast agent (1), the hardness of the object represented by the elastic modulus and degree of curing can be obtained.

[0051] The contrast agent for visualizing such hardness has no particular limitation on structure as long as it is a compound capable of emitting light from two or more different excited states depending on the surrounding environment.

[0052] [Chemical formula 1]

[0053]

[0054] Such a pigment can be synthesized with reference to the above-mentioned Chem. Sci., 2020, 11, p7525.

[0055] For the above-mentioned contrast agent (1), in an environment where structural relaxation is difficult, that is, when the elastic modulus of the object is high, the peak of the luminous intensity is near a wavelength of 488nm. On the other hand, in an environment where structural relaxation is easy, that is, when the elastic modulus of the object is low, the peak of the luminous intensity is near a wavelength of 590nm. Therefore, according to the elastic modulus of the object, the ratio (F488 / F590) of the luminous intensity (F488) at a wavelength of 488nm and the luminous intensity (F590) at a wavelength of 590nm changes. That is, the ratio (F488 / F590) of the luminous intensity (F590) as spectral characteristic information can be associated with the elastic modulus.

[0056] The method of correlating the ratio of luminous intensities and the elastic modulus will be described in more detail.

[0057] Figure 2A shows images captured by a conventional camera at different temperatures when the temperature of the object containing the contrast agent (1) is changed, Figure 2B It represents the spectral spectrum obtained by averaging the spectral spectrum of a specific area captured by the hyperspectral camera at each temperature. Figure 2B In the figure, the horizontal axis represents the wavelength (nm) and the vertical axis represents the luminous intensity (-). Figure 3 This is a graph showing the behavior of the ratio of the luminous intensity at a wavelength of 488nm to the luminous intensity at a wavelength of 590nm (F488 / F590) obtained from the spectral spectrum and the behavior of the elastic modulus (Log value) measured separately when the temperature of the object is changed. Figure 3 In the graph, the horizontal axis represents the temperature of the object (° C.), the left vertical axis represents the ratio of luminous intensity (F488 / F590), and the right vertical axis represents the Log value of the elastic modulus.

[0058] exist Figure 2A In the figure, it is shown that as the temperature rises, the luminescent color changes to blue (around 23 to 70°C), pink (80 to 110°C), and orange (120 to 150°C). Figure 2BIn the figure, it can be seen that as the temperature rises, the peak of the luminous intensity shifts to the long wavelength side. That is, as the temperature rises, the ratio of the long wavelength component (red luminous component) to the short wavelength component (blue luminous component) increases, that is, the ratio of the luminous intensity at a wavelength of 488nm to the luminous intensity at a wavelength of 590nm (F488 / F590) decreases (refer to Figure 3 ). Moreover, it can be seen that the temperature change behavior of the ratio of luminous intensities is roughly the same as (corresponds to) the temperature change behavior of the elastic modulus actually measured. Therefore, the ratio of luminous intensities can be associated with the elastic modulus. By performing this operation for each pixel on the two-dimensional coordinates, the two-dimensional data of the ratio of luminous intensities can be associated with the two-dimensional data of the elastic modulus. Thus, as data associated with the elastic modulus, the two-dimensional data of the ratio of luminous intensities can be used in the prediction process described later.

[0059] In addition, as compounds whose luminescent behavior changes according to the water content and polarity in the object, there are lyotropic pigments. Lyotropic pigments are pigments whose luminescent wavelength and absorption wavelength change according to the water content and polarity of the surrounding objects and solvents. This is understood to be that the energy difference with the base state changes due to the structural stabilization or destabilization of the excited state of the pigment according to the type / composition of the solvent, so that the luminescent wavelength corresponding to the energy difference changes. In other words, it means that by observing the luminescent wavelength, the type / composition of the surrounding objects and solvents that stabilize or destabilize the contrast agent can be indirectly visualized.

[0060] As a contrast agent that visualizes such water content and polarity, the structure is not particularly limited as long as the luminescence wavelength and absorption wavelength of the pigment change according to the surrounding object, the water content and polarity of the solvent. Examples of such contrast agents include, for example, squaric acid pigments such as contrast agent (2).

[0061] [Chemical formula 2]

[0062]

[0063] Such a pigment can be synthesized with reference to the above-mentioned WO2020 / 171199 publication.

[0064] In addition, the wavelength of the light (excitation light) irradiated to the object is appropriately selected according to the type of contrast agent, the type of object, etc. For example, when the contrast agent is a compound that can be excited by visible light, visible light is used as the excitation light. On the other hand, when the contrast agent is a compound that can be excited by ultraviolet light, ultraviolet light is used as the excitation light.

[0065] The other of the two or more data preferably includes film thickness information, because this data is not dependent on the type of object or bonding method but has a deep relationship with the bonding state.

[0066] In this embodiment, if Figure 1 As shown, for example, an image (image 1) is acquired by a hyperspectral camera under the condition of irradiating light of a predetermined wavelength (step S11). Then, data (data 1) including a ratio of luminous intensity (F488 / F590) associated with the elastic modulus is acquired from the image (step S12).

[0067] In addition, a reflection spectrum image (image 2) is acquired by the reflection spectroscopic film thickness meter (step S13). Then, data (data 2) including the film thickness is acquired from the reflection spectrum image (step S14).

[0068] (Prediction process)

[0069] In the prediction process (step S15), the bonding state between the object and the adherend is analyzed and predicted using the two or more data acquired in the data acquisition process (steps S11 to S14). Rather than using a single data to perform such analysis and prediction, the analysis and prediction are performed on two or more data, thereby improving the accuracy of the analysis and prediction.

[0070] The adhesion state can be analyzed and predicted by any method. For example, the adhesion state of the object may be analyzed and predicted by comparing two or more data acquired over time for predetermined physical property value information.

[0071] In addition, data of the object in a standard state (for example, when the adhesion state is good) and in a predetermined state (for example, when the adhesion state is poor and peeling occurs) may be acquired in advance, and these data may be compared with the data acquired in the above data acquisition step to analyze and predict the adhesion state of the object. The predetermined state may be, for example, when the adhesion state is good or when poor adhesion occurs.

[0072] In addition, the analysis and prediction of the bonding state can also be based on a prediction model (learned model) generated in advance by machine learning, etc. In the case of performing the analysis based on the learned model, by applying two or more data obtained in the above-mentioned data acquisition step to the learned model, it is possible to determine (predict) what bonding state the object is in based on the accumulated data, etc.

[0073] Figure 4 is the prediction process indicating the bonding state ( Figure 1 A flowchart of an example of step S15).

[0074] In machine learning, for example, the same process as the above-mentioned data acquisition process is performed multiple times. And, based on this, multiple prediction models are constructed. And, by combining the results of multiple prediction models, a learning model (adhesion prediction state algorithm) that can predict information related to the adhesion state of the object (for example, peeling force, etc.) is prepared.

[0075] The above prediction model can be constructed by performing machine learning using two or more data as explanatory variables and a physical property representing the adhesion state of the object (e.g., peeling force) as a target variable, etc., when the state of the object is determined in advance. As the explanatory variable, the same data as the data acquired in the above data acquisition step (S11 to S14) can be used. As the target variable, it can be appropriately selected according to the purpose of the analysis, and a variable related to the adhesion state of the object (e.g., peeling force, cross-cut test, pencil hardness test, etc.) can be used.

[0076] Machine learning can be supervised learning or unsupervised learning. In addition, supervised learning refers to a learning method that learns the "relationship between input and output" from learning data with correct answer labels. Unsupervised learning refers to a learning method that learns the "structure of a data group" from learning data without correct answer labels.

[0077] In addition, machine learning can also be reinforcement learning, deep learning, or deep reinforcement learning. In addition, reinforcement learning refers to a learning method that learns the "best series of actions" through trial and error. Deep learning refers to a learning method that gradually learns the features contained in the data more deeply (deeply) from a large amount of data. Deep reinforcement learning refers to a learning method that combines reinforcement learning and deep learning.

[0078] Machine learning can apply general analysis methods (algorithms). For example, machine learning can apply a prediction model constructed by an analysis method selected from linear regression (multiple regression analysis, partial least squares (PLS) regression, LASSO regression, Ridge regression, principal component regression (PCR), etc.), random forest, decision tree, support vector machine (SVM), support vector regression (SVR), neural network, discriminant analysis, etc.

[0079] In the prediction process of the bonding state, the above-made learned model is used to analyze and predict the bonding state. First, the learned model is read in (step S21). In this embodiment, for example, a regression learning model is read in. The explanatory variables of the regression are set to two or more data obtained in the above-mentioned data acquisition process, and the target variable is set to the peeling force.

[0080] Next, two or more data (for example, data including the ratio of luminous intensity and data including film thickness) input as explanatory variables are extracted from the data set obtained in the above data acquisition step. Then, the extracted data is input into the explanatory variables of the regression equation of the learned model for each pixel (step S22). Then, the peeling force is predicted for each pixel (step S23) and output as the target variable (step S24). This step is performed for each of the two or more data.

[0081] Next, the peeling force of each pixel output is plotted as a two-dimensional coordinate (step S24). Moreover, the region is divided according to the level of the peeling force, and each region is divided by color, and visualized on the two-dimensional coordinate. For example, the region where the peeling force exceeds the specified threshold is set as the NG region (region where poor adhesion occurs), and the region where the peeling force does not exceed the threshold is set as the OK region (region where poor adhesion does not occur), and the color is divided and displayed. Thus, on the two-dimensional coordinate, it is possible to visualize and predict which part of the object has poor adhesion.

[0082] Furthermore, when it is difficult to determine a threshold value and to clearly define a boundary, it is preferable to clarify the contour of the acquired image using a HOG (Histograms of Oriented Gradients) feature.

[0083] The "HOG feature" is a feature of the local image gradient after histogramming. By obtaining the HOG feature and detecting the gradient of the peeling force, the contour can be clarified. The HOG feature can be calculated using various well-known papers and Japanese Patent Publication No. 2018-36689 as references. In addition, the above-mentioned machine learning method can also be used as the algorithm for clarifying the boundary.

[0084] Furthermore, when the peeling force of the object is actually measured in contrast to the prediction step S15, the result of the actual peeling force measurement is compared with the prediction result, and the compared result is added to the teacher data of the learned model, thereby further improving the prediction accuracy of the learned model.

[0085] (Effect)

[0086] Conventionally, separate data is acquired and the adhesion state of the object is predicted from the acquired data, so the prediction accuracy is insufficient.

[0087] In contrast, in the present embodiment described above, two or more data are acquired, and the adhesion state of the object is predicted using the acquired two or more data. This enables multivariate analysis compared to the prior art, and thus improves prediction accuracy.

[0088] In addition, by preferably setting at least one of the two or more data to include spectral characteristic information (e.g., the ratio of luminous intensities), the data required for predicting the bonding state can be acquired in a non-contact and real-time manner. Furthermore, preferably, the prediction process is performed using a learned model, so that the bonding state can be predicted with higher accuracy.

[0089] In addition, in the above embodiment, after two or more data are acquired in the data acquisition process (steps S11 to S14), the two or more data are used to predict the bonding state in the prediction process (step S15), but this is not limited to this, and the prediction process can also be performed each time one or two or more data are acquired.

[0090] Figure 5 FIG. 4 is a flowchart showing a method for predicting an adhesion state according to another embodiment of the present invention. Figure 5 In the process, a spectroscopic image (image 1) is obtained (step S31), and data (data 1) including a ratio of luminous intensity (F488 / F590) associated with the elastic modulus is obtained therefrom (step S32). Then, the bonding state prediction process is performed using the obtained data (step S33).

[0091] Next, a reflection spectrum image (image 2) is obtained (step S34), and data (data 2) including film thickness is obtained therefrom (step S35). Then, the bonding state prediction process is performed using the obtained data (step S36). In this way, the prediction process may be performed each time one or more data are obtained.

[0092] also, Figure 5 Steps S31, S32, S34 and S35 in Figure 1 The steps S11, S12, S13 and S14 correspond to each other. Figure 5 The steps S33 and S36 in the above are combined into a step Figure 1 Corresponding to step S15.

[0093] In the above embodiment, data including the ratio of luminous intensity or elastic modulus and data including film thickness are obtained as two or more data, but the present invention is not limited to this, and appropriate data can be obtained according to the type of object, the type of adherend, and the bonding method.

[0094] 2. Prediction system of bonding state

[0095] The method for predicting the adhesion state of the present embodiment can be performed by the following system for predicting the adhesion state. In addition, the system for performing the method for predicting the adhesion state is not limited to the following system.

[0096] Figure 61 is a schematic diagram showing the structure of the adhesion state prediction system 100 according to the present embodiment. Figure 6 As shown, the prediction system 100 of this embodiment includes an imaging device 110 , a processing device 120 , and a display unit 130 .

[0097] (Camera device)

[0098] The imaging device 110 captures an image showing a light-emitting state of an object when irradiated with light. The imaging device 110 includes a light source 111 and an imaging unit 112 .

[0099] The light source 111 is not particularly limited as long as it is a device that can irradiate light of a specified wavelength to an object. As the light source 111, a lamp with a wide range of wavelengths can be applied, including light sources in the ultraviolet, visible range, near infrared, and infrared regions in this example. For example, xenon lamps, halogen lamps, white LED lamps, near-infrared hyperspectral imaging lighting (LDL-222X42CIR-LACL manufactured by CCS Co., Ltd., etc.), laser-excited white light sources that can radiate light in the wavelength range from deep ultraviolet to near infrared (XWS-65 manufactured by KLV Co., Ltd., etc.), etc. can be used. When using fluorescent luminescent materials, it is preferred to include a light source in the ultraviolet region, and when using infrared pigments, etc., it is preferred to include a light source of infrared light. In addition, when a specific material such as a contrast agent is excited to emit light, a light source with a sharp waveform such as an LED light source can enhance the spectrum unique to the contrast agent, so it is preferred. The shape of the light source may be a normal point light source, and when installed in a production line or the like, it is preferable to use line lighting (high-brightness spotlight line lighting manufactured by CCS, such as LDL-222X42CIR-LACL).

[0100] The photographing unit 112 is not particularly limited as long as it is a unit that can photograph the state of light reflected or released by the object receiving light from the light source 111, and is appropriately selected according to the type of data to be acquired. For example, the photographing device 110 may also be a monochrome camera, a color camera, an infrared camera, a multispectral camera, a hyperspectral camera, etc. The image captured by the photographing device 110 is output to the data acquisition unit 121.

[0101] In addition, multispectral cameras and hyperspectral cameras are cameras that can shoot at more wavelengths than ordinary cameras, and are cameras with high spectral resolution and spatial resolution. Therefore, they can evaluate the object quantitatively at multiple points / wide ranges through a single measurement, so they are preferred. The wavelength resolution is preferably 50nm or less, more preferably 10nm or less, and further preferably 5nm or less. However, if the resolution is increased excessively, the amount of data processing increases, and the load / processing time on the data processing unit increases, so the minimum required resolution is preferred.

[0102] In addition, multispectral cameras and hyperspectral cameras include area type (snapshot type) and line type. The area type is preferred when observing the shape of a component or an individual part, and the line type is preferred when observing the shape of a roll.

[0103] As hyperspectral cameras capable of measuring the visible light region, there are SpecimIQ and FX-10 from Specim, and NH series from EBA JAPAN. As hyperspectral cameras capable of measuring the near-infrared region, there are FX-17 and SW-IR from Specim, PikaNIR-320 from Resonon, SWIR-640 from HySpex, SNAPSCAN-SWIR from Imec, SIS-IR series and SIS-SWIR series from EBA JAPAN. As hyperspectral cameras capable of measuring the mid-infrared region, there are FX-50 and MW-IR from Specim. As hyperspectral cameras capable of measuring the far-infrared region, there are LW-IR from Specim, but are not limited to these.

[0104] (Processing device)

[0105] The processing device 120 has: a data acquisition unit 121 for acquiring data of two or more objects from the above-mentioned image; a storage unit 122 for storing the two or more acquired data; and a bonding state prediction unit 123 for predicting the bonding state between the object and the adherend using the two or more acquired data.

[0106] In the data acquisition unit 121, the above-mentioned data acquisition process is performed. That is, the two-dimensional coordinate information and the physical property value information linked thereto are acquired. In the present embodiment, the data acquisition unit 121 comprises: an image acquisition unit 124 that acquires an image captured by the imaging device 110; and a processing unit 125 that acquires two or more of the above-mentioned data (including the data of the physical property value information on the two-dimensional coordinates) from the image.

[0107] The image acquisition unit 124 may be any unit that can acquire an image captured by the imaging device 110 or an image captured by an external device (not shown).

[0108] The processing unit 125 may be any unit that can acquire data of an object from an image acquired by the image acquisition unit 124. For example, the processing unit 125 can acquire data including a ratio of luminous intensities (F488 / F590) from a spectral image acquired by the image acquisition unit 124. In addition, depending on the type of image acquired by the image acquisition unit 124, the processing unit 125 may not be required, or processing by the processing unit 125 may not be performed. For example, temperature information is directly obtained from a temperature distribution image obtained by infrared thermal imaging, and therefore processing by the processing unit 125 is not required. In addition, film thickness information is directly obtained from reflective spectrum data obtained by a reflective spectroscopic film thickness meter, and therefore processing by the processing unit 125 is not required.

[0109] The storage unit 122 may be any unit as long as it can store two or more data acquired by the processing unit 125 .

[0110] In the adhesion state prediction unit 123, the above prediction process is performed. The adhesion state prediction unit 123 can be any device as long as it can analyze the data obtained by the data acquisition unit 121 (for example, the processing unit 125). For example, data containing reference property value information obtained separately can also be read out, and the reference data can be compared with the data obtained from the data acquisition unit 121 (for example, the processing unit 125) to analyze and predict the adhesion state. In addition, the adhesion state prediction unit 123 can also predict the adhesion state based on the learned model. Specifically, the adhesion state prediction unit 123 can also read the learned model from the storage unit 122 or an external storage device (not shown), input the data obtained by the data acquisition unit 121 (for example, the processing unit 125) into the learned model, and perform calculations. Then, the calculation result, that is, the prediction result of the adhesion state, is output.

[0111] As the processing device 120, a general computer (general-purpose computer) can be used, which has a storage unit such as a hard disk drive (HDD), a solid state drive (SSD), and a read-only memory (ROM) for storing programs, data, etc., and a central processing unit (CPU) for executing programs, performing calculations, etc. In addition, the computer may also have input units such as a keyboard and a mouse, and output units such as a monitor and a printer.

[0112] (Display)

[0113] The display unit 130 may be any unit as long as it can display the result predicted by the adhesion state prediction unit 123. The display unit 130 may be an output unit such as a monitor or a printer. The display unit 130 may be a unit integrally formed with the processing device 120.

[0114] (Effect)

[0115] The prediction system 100 of this embodiment includes the data acquisition unit 121 that acquires two or more data on the object and the adhesion state prediction unit 123 that predicts the adhesion state of the object using the acquired two or more data. This allows multivariate analysis compared to the conventional method, thereby improving prediction accuracy.

[0116] In the above embodiment, the prediction system 100 includes the imaging device 110, but may not include the imaging device 110. For example, the prediction system 100 may be configured so that the image acquisition unit 124 reads an image representing a light emission state acquired separately by an external imaging device (not shown).

[0117] In addition, in the above embodiment, an example is shown in which the data acquisition unit has an image acquisition unit 124 that simultaneously acquires two-dimensional coordinate information and physical property value information, but this is not limited to this, and it may also be a two-dimensional coordinate information acquisition unit that separately acquires two-dimensional coordinate information and the corresponding physical property value information.

[0118] In the above embodiment, the prediction system 100 includes the storage unit 122, but may not include the storage unit 122. For example, the prediction system 100 may be configured to directly input two or more data acquired by the processing unit 125 to the adhesion state prediction unit 123 from the processing unit 125, or may be configured to read them from an external storage device (not shown).

[0119] In the above embodiment, the prediction system 100 includes the display unit 130 , but the display unit 130 may not be included, and the result output from the adhesion state prediction unit 123 may be displayed on an external display device (not shown).

[0120] 3. Prediction procedure of bonding state

[0121] The method for predicting the adhesion state of the present embodiment can be performed by the following adhesion state prediction program.

[0122] That is, the program for predicting the adhesion state of the present embodiment is for causing a computer to execute the above-mentioned data acquisition step (for example, Figure 1 Steps S11 to S14), a bonding state prediction step (e.g. Figure 1 The content of each step of the prediction program is the same as the content of each step of the prediction method.

[0123] The prediction program may be provided by being stored in a recording medium such as a DVD or a USB memory, or may be stored in a server device on the network so as to be downloadable via the network.

[0124] 4. Method for manufacturing bonded article

[0125] (Implementation Method 1)

[0126] The above-mentioned method for predicting the bonding state can be applied to the manufacturing process of various devices and their components. Examples of devices and their components include displays, polarizing plates, touch panels, optical films, etc. The above-mentioned method for predicting the bonding state can be applied, for example, to the process of bonding two adherends together via a thermocompression bonding film in the manufacturing method of the device.

[0127] Figure 7 This is a flowchart showing an example of a method for producing a bonded product according to the present embodiment.

[0128] like Figure 7 As shown, first, a process of preparing a thermocompression bonding film (object) is performed (preparation process, step S41), and a heat treatment is performed to melt the thermocompression bonding film (heat treatment process, step S42). Next, data containing physical property value information is obtained for the thermocompression bonding film after heating and melting (step S43), and the bonding state of the thermocompression bonding film / glass film interface when the thermocompression bonding film after heating and melting is bonded to the glass film (adhesive body) is predicted (bonding state prediction process, step S44). Then, the predicted result of the bonding state is visualized (visualization process, step S45), and it is determined whether peeling occurs (judgment process, step S46). If it is determined that peeling does not occur, it is bonded to the glass film to obtain an adhesive (bonding process, step S47). On the other hand, if it is determined that peeling occurs, the heating treatment conditions are modified again (adjustment process, step S48), and the heating treatment process is performed again (heat treatment process, step S42). Below, each process is described.

[0129] 1) Preparation process (step S41)

[0130] First, a thermocompression bonding film is prepared. The thermocompression bonding film can also be prepared by imparting a thermocompression bonding material to a substrate such as a resin film, or a film formed by imparting a thermocompression bonding material to a substrate in advance can be used. The method of imparting the thermocompression bonding material to the substrate is not particularly limited, and can be a coating method or a melt extrusion method. In addition, in the present embodiment, a resin film is used as the substrate, but other substrates can also be used depending on the application.

[0131] In this embodiment, the thermocompression bonding film preferably includes a contrast agent suitable for obtaining data including elastic modulus (e.g., the contrast agent (1) described above, with an excitation wavelength of 365 nm). The reason why the contrast agent is preferably included is that, in order for the thermocompression bonding film to be bonded to an adherend with sufficient adhesive force, it is very important that the hot melt adhesive material used as the adhesive of the thermocompression bonding film, the fluidity of the hot melt adhesive material when heated and melted, and the elastic modulus when bonded after cooling are within an appropriate range, which is particularly effective information when predicting the bonding state.

[0132] 2) Heat treatment process (step S42)

[0133] Next, the thermocompression bonding film is subjected to a heating treatment. Thus, the thermocompression bonding film is heated and melted to be in a state that facilitates bonding. The heating temperature can be set to, for example, near the glass transition temperature Tg of the thermocompression bonding film.

[0134] 3) Adhesion state prediction process (steps S43 to S44)

[0135] Next, the method for predicting the adhesion state of the present invention is implemented. Figure 1 The steps shown implement a method for predicting the bonding state.

[0136] Specifically, an image (image 1) showing the luminous state when the heat-compression-bonded film after heating and melting is irradiated with light of an excitation wavelength of 365 nm is obtained by a hyperspectral camera. Then, data (data 1) including the ratio of luminous intensity (F488 / F590) associated with the elastic modulus is obtained. The spectral image can also be obtained multiple times over time.

[0137] In addition, the reflection spectrum data (image 2) of the heat-pressed film after heating and melting is obtained by the reflection spectrophotometer, and the data including the film thickness (data 2) is obtained. The reason for obtaining the data including the film thickness is that the film thickness of the adhesive layer is one of the important factors that determine the adhesive force. Even if the above-mentioned fluidity and elastic modulus are within the appropriate range, if the film thickness is not within the appropriate range, sufficient adhesive force cannot be reflected, so the film thickness is also effective information when predicting the bonding state.

[0138] Furthermore, a temperature distribution image (image 3) of the heat-compression-bonded film after heating and melting is obtained by infrared thermal imaging, and temperature data (data 3) is obtained (step S43). The reason for obtaining the temperature data is as follows. For example, when the fluidity of the adhesive visualized in the contrast agent (1) does not meet expectations, if the film thickness meets expectations, it is believed that the reason is that the temperature is not sufficiently high, the temperature distribution is uneven, etc. Therefore, when understanding why the predicted result of the bonding state is like that, temperature is important information.

[0139] Then, using the acquired data 1, 2, and 3, the bonding state of the thermocompression bonding film / glass film interface when the glass film (adherend) is bonded is predicted (step S44). Figure 4 As shown, the above data 1, 2, and 3 are input to the explanatory variables of the learned model, and the peeling force is output as the target variable. Specifically, the peeling force is output for each position on the two-dimensional coordinates.

[0140] 4) Visualization process (step S45)

[0141] Next, the results output in the above-mentioned adhesion state prediction step are visualized. There is no particular limitation on the visualization method. For example, at each position on the two-dimensional coordinates, the color is divided into the part where the peeling force exceeds the threshold (peeling force NG area) and the part where it does not exceed the threshold (peeling force OK area) and displayed. In this way, the adhesion state of the thermocompression bonding film and the glass film after the lamination process (including after the endurance test) can be predicted in advance.

[0142] 5) Judgment process (step S46)

[0143] Next, based on the visualized data, it is determined whether peeling has occurred. The determination method is not particularly limited, and it can be performed, for example, by comparing with the data already acquired. Then, in the determination process, if it is determined that peeling has not occurred, the bonding process (step S47) is performed. On the other hand, if it is determined that peeling has occurred, the adjustment process (step S48) is performed.

[0144] 6) Lamination process (step S47)

[0145] If it is determined in the determination step that no peeling has occurred, a glass film is bonded to the heated and melted thermocompression bonding film. Thus, a bonded product of the thermocompression bonding film and the glass film (a bonded product of the resin film and the glass film bonded via the thermocompression bonding material) can be obtained.

[0146] 7) Adjustment process (step S48)

[0147] If peeling occurs in the above-mentioned judgment process, the heating treatment conditions in the heating treatment process (step S42) are adjusted. For example, the temperature and time to be heated are set based on the distribution of the peeling force. Then, the heating process is returned to the heating treatment process (step S42), and the heating process is performed again under the conditions set in the adjustment process (step S42). This step is repeated in the judgment process until the NG area of ​​the peeling force is not detected.

[0148] As described above, the method for manufacturing an adhesive product of the present embodiment applies the method for predicting the bonding state of the present invention to a thermocompression bonding film (an object having adhesiveness), and includes: a step of predicting the bonding state between the thermocompression bonding film and the glass film when the glass film (adherend) is bonded together; and a step of adjusting the processing conditions of the object based on the prediction result. Thus, the bonding state when the glass film is bonded together can be predicted, and the adhesive product can be manufactured, thereby improving the manufacturing efficiency.

[0149] Furthermore, in the above-mentioned embodiment, the Figure 1 The example of performing the bonding state prediction process (steps S43 to S44) is shown in the steps shown in FIG. Figure 5 Follow the steps shown.

[0150] Furthermore, in the above-described embodiment, an example is shown in which a step of visualizing the result outputted in the adhesion state prediction step (visualization step) is performed, but the step may be performed as necessary and need not be performed.

[0151] In addition, in the above-mentioned embodiment, the bonding state between the thermocompression bonding film and the adherend is predicted, but it is not limited to this. For example, the bonding state between the curable material or the adhesive material and the adherend can also be predicted. Therefore, the heating treatment process (step S42) of the above-mentioned embodiment only needs to be a process corresponding to the type of the object. For example, when the object is a photocurable material, the treatment process can be a light irradiation process.

[0152] (Implementation Method 2)

[0153] The method for predicting the adhesion state of the present invention can be applied to, for example, a process of forming an insulating protective layer such as a solder resist for circuit protection on the surface of a printed wiring board in a method for manufacturing a wiring substrate such as a printed wiring board.

[0154] In the solder resist forming process, first, a photocurable material is applied to a printed wiring board (coating process). Then, the obtained coating film is irradiated with light and cured (curing process). Thus, a solder resist containing a cured product of the photocurable material is formed on the surface of the printed wiring board.

[0155] The photocurable material generally comprises a photocurable compound and a photopolymerization initiator. The photocurable compound is preferably a free radical curable compound. In this embodiment, the photocurable material preferably further comprises: a contrast agent suitable for obtaining data including hardness (e.g., the contrast agent (1) described above, with an excitation wavelength of 365 nm), and a contrast agent suitable for obtaining data including polarity (e.g., the contrast agent (2) described above, with an excitation wavelength of 660 nm).

[0156] The reason why it is preferred to include a contrast agent (1) that can visualize hardness is that the hardness in an appropriate range is required for the photocurable material to maintain sufficient adhesion even in the subsequent steps (development step, welding step, etc.), and therefore the hardness is effective information for predicting the adhesion state. According to the contrast agent (1), not only the hardness but also whether the degree of curing is within an appropriate range, that is, whether the curing reaction is completed, can be visualized, which is preferred.

[0157] The reason why the contrast agent (2) capable of visualizing polarity is preferably included is that electronic properties such as polarity caused by functional groups of monomers contained in the photocurable material are one of the important factors affecting the adhesiveness to an adherend such as a printed wiring board.

[0158] Moreover, if Figure 1 As shown, a method for predicting the bonding state is performed.

[0159] Specifically, in the coating process, a spectral image (image 1) showing the luminous state when the coating film is irradiated with light of an excitation wavelength of 660 nm is obtained by a hyperspectral camera. In addition, the ratio of the luminous intensity at a wavelength of 800 nm to the luminous intensity at a wavelength of 750 nm (F800 / F750) associated with polarity can be obtained (data 1).

[0160] In addition, during the curing process, a spectral image (image 2) showing the luminous state when the coating film is irradiated with light having an excitation wavelength of 365 nm is obtained by a hyperspectral camera. In addition, data (data 2) including the ratio of the luminous intensity at a wavelength of 430 nm to the luminous intensity at a wavelength of 600 nm (F430 / F600) which is associated with hardness is obtained.

[0161] Then, using the acquired data 1 and 2, a bonding state prediction step is performed at each timing after the coating step and after the curing step. Thus, the final (including after the endurance test) bonding state between the cured product of the photocurable material, that is, the solder resist and the printed wiring board can be predicted in advance.

[0162] (Implementation 3)

[0163] The method for predicting the adhesion state of the present invention can be applied to a method for producing a hard coating film, for example, a substrate for a flexible display such as an organic EL display, a barrier film for preventing water penetration, and the like, and a hard coating layer having a light-transmitting resin film and a cured product containing a photocurable material.

[0164] In the manufacturing process of such a hard coating film, a photocurable material is coated on a resin film (coating process). Then, the coated photocurable material is irradiated with light and cured (curing process). Thus, a hard coating layer containing a cured product of the photocurable material is formed.

[0165] In this embodiment, the photocurable material may include a photocurable compound and a photopolymerization initiator as described above. The photocurable compound is preferably a free radical curable compound. In this embodiment, the photocurable material preferably further includes a contrast agent suitable for obtaining data including the degree of curing (e.g., the contrast agent (1) described above, with an excitation wavelength of 365 nm).

[0166] Moreover, if Figure 5 As shown, a method for predicting the bonding state is performed.

[0167] Specifically, in the coating process, a spectral image (image 1) showing the luminescence state when the coating film is irradiated with light of an excitation wavelength of 365 nm is obtained by a hyperspectral camera, thereby obtaining data (data 1) including luminescence intensity at a wavelength of 600 nm associated with coating unevenness.

[0168] That is, in order for the hard coating layer to adhere to the resin film with sufficient adhesion, the first major premise is whether the hard coating layer is uniformly formed. Therefore, by using a contrast agent (1), it is possible to visualize whether the amount of coating film attached and its distribution are within an appropriate range (whether the uneven coating is suppressed to a certain level or below), that is, whether the minimum necessary conditions for adhesion are met, so that the first step of predicting the adhesion state described later can be performed. In this way, by dividing and executing the prediction of the adhesion state according to each step, if the adhesion abnormality can be predicted in the early stage of the process, it will lead to shortened time and reduced loss, which is preferred.

[0169] In addition, in the curing process, a spectral image (image 2) showing the luminous state when the coating film is irradiated with light of an excitation wavelength of 365 nm is obtained by a hyperspectral camera. Thus, data (data 2) including the ratio of luminous intensity (F430 / F600) associated with the degree of cure is obtained. By visualizing the degree of cure in an appropriate range, i.e., the completion of the curing reaction, by using a contrast agent (1), the second step of predicting the bonding state described later can be performed. The second step can perform a prediction with better accuracy than the prediction in the first step.

[0170] In addition, during the curing process, the reflection spectrum data of the coating film is obtained by a reflection spectroscopic film thickness meter (image 3). Thus, data including the film thickness is obtained (data 3). As described in embodiment 1, the film thickness information is also an important factor in predicting the bonding state, and is preferably obtained when improving the prediction accuracy.

[0171] Then, using the acquired data 1, a bonding state prediction step ( Figure 5 Adhesion state prediction 1, step S33). Thus, the adhesion state between the hard coating layer and the resin film after the coating process can be predicted. In addition, using the acquired data 2 and data 3, the adhesion state prediction process is performed after the curing process ( Figure 5 The bonding state prediction 2, step S33). Thus, the final bonding state between the hard coating layer and the resin film can be predicted.

[0172] (Implementation 4)

[0173] The method for predicting the adhesive state of the present invention can also be used for evaluating an adhesive film in a method for producing an adhesive film, for example.

[0174] Specifically, after the adhesive material is applied on the resin film (substrate) (coating step), the applied adhesive material is dried (drying step), and then the obtained adhesive film is evaluated.

[0175] The adhesive material includes an adhesive as a main agent. In the present embodiment, the adhesive material preferably further includes a contrast agent suitable for obtaining data including uneven coating (such as the above-mentioned contrast agent (1), excitation wavelength 365nm), and a contrast agent suitable for obtaining data including residual solvent amount (such as the above-mentioned contrast agent (2), excitation wavelength 660nm). That is, whether the adhesive material layer formed on the resin film can exert sufficient adhesive force to the adherend depends on whether the adhesive material layer is uniformly formed with an appropriate film thickness range. Thus, it is preferred to include a contrast agent (1) that can visualize uneven coating, and it is preferred to obtain film thickness data at the same time. In addition, if there is residual solvent in the film, it may become a factor that reduces adhesive force, so it is preferred to include a contrast agent (2) that can visualize the residual solvent amount.

[0176] Moreover, if Figure 1 As shown, a method for predicting the bonding state is performed.

[0177] Specifically, in the coating process, a spectral image (image 1) showing the luminescence state when the coating film is irradiated with light of an excitation wavelength of 365 nm is obtained by a hyperspectral camera, thereby obtaining data (data 1) including luminescence intensity at a wavelength of 600 nm associated with coating unevenness.

[0178] In addition, in the coating process, the reflection spectrum data of the coating film is acquired by a film thickness reflection spectrometer (Image 2). Thus, data including the film thickness is acquired (Data 2).

[0179] In the drying process, a spectral image (image 3) showing the luminescence state when the coating film is irradiated with light of an excitation wavelength of 660 nm is obtained by a hyperspectral camera. Thus, data (data 3) including luminescence peak shift (F750-800) associated with the amount of residual solvent is obtained.

[0180] Then, a step of predicting the adhesion state when the adhesive film is used in the laminating step is performed using the acquired data 1, data 2, and data 3. Thus, the adhesion state between the adhesive material and the adherend after the laminating step can be predicted.

[0181] (Implementation method 5)

[0182] The method for predicting the adhesion state of the present invention can be applied to, for example, a method for producing a polarizing plate. A polarizing plate includes a polarizer, a light-transmitting resin film, and a cured product of a photocurable material disposed therebetween.

[0183] In the manufacturing process of such a polarizing plate, for example, a process of coating a photocurable material on a polarizer (coating process). Then, a light-transmitting resin film is bonded to the coated photocurable material (bonding process). Then, the bonded resin film is irradiated with light to cure the photocurable material (curing process). In this way, a polarizing plate can be manufactured.

[0184] The photocurable material is the same as above, comprising a photocurable compound and a photopolymerization initiator. The photocurable compound is preferably a cationic curable compound. In the present embodiment, the photocurable material preferably further comprises a contrast agent suitable for obtaining data containing a degree of cure (e.g., the contrast agent (1), with an excitation wavelength of 365 nm), and a contrast agent suitable for obtaining data containing water content (e.g., the contrast agent (2), with an excitation wavelength of 660 nm). The reason why the contrast agent (2) that can visualize the water content is preferably included is because: the moisture contained in the cationic curable compound may become a factor causing curing hindrance, and therefore sometimes becomes important information when understanding the cause of insufficient curing. In addition, if the contrast agent (1) that can visualize the degree of cure is included, it is possible to judge whether the degree of cure is within an appropriate range, i.e., whether the curing reaction is completed, and therefore it is preferred. In this way, the reason for visualizing the curing reaction at an appropriate level is because: if the curing is insufficient, the target adhesion cannot be obtained, and therefore the degree of cure is effective information when predicting the bonding state.

[0185] Moreover, if Figure 1 As shown, a method for predicting the bonding state is performed.

[0186] Specifically, in the curing process, a spectral image (image 1) showing the luminous state when the coating film is irradiated with light of an excitation wavelength of 365 nm is obtained by a hyperspectral camera. In addition, data (data 1) including the ratio of the luminous intensity at a wavelength of 430 nm to the luminous intensity at a wavelength of 600 nm (F430 / F600) associated with the degree of curing is obtained.

[0187] In addition, during the lamination process, a spectral image (image 2) showing the luminescence state when the cured product is irradiated with light having an excitation wavelength of 660 nm is acquired by a hyperspectral camera. In addition, data (data 2) including luminescence peak shift (F750-800) associated with water content is acquired.

[0188] Then, using the acquired data 1 and data 2, a bonding state prediction step is performed after the polarizing plate is produced and before it is extracted for inspection. This makes it possible to predict the final bonding state of the polarizing plate (including after the endurance test).

[0189] (Implementation 6)

[0190] In the method for predicting the adhesion state of the solder resist to the printed wiring board described in Embodiment 2, the adhesion state can be predicted by acquiring near-infrared reflection spectra (eg, 1000 nm to 2700 nm) of the photocurable material before and after curing using a hyperspectral camera.

[0191] Moreover, if Figure 1 As shown, a method for predicting the bonding state is performed.

[0192] Specifically, in the coating process, a spectral image (image 1) showing the near-infrared absorption spectrum when the coating film is irradiated with halogen lamp as a light source is acquired by a hyperspectral camera, and data (data 1) associated with molecular structure information of the photocurable material before curing is acquired.

[0193] Next, in the curing step, a spectral image (image 2) showing the near-infrared absorption spectrum when the coating film is irradiated with halogen lamp light as a light source is acquired by a hyperspectral camera, and data (data 2) associated with molecular structure information of the cured photocurable material is acquired.

[0194] Infrared light is absorbed by the vibration and rotation of molecules, and its energy varies depending on the chemical structure. Therefore, if infrared light is measured, information related to the chemical structure and the state of the molecule can be obtained. Specifically, data 1 and data 2 can be obtained from the reflected light obtained by the hyperspectral camera as a near-infrared absorption spectrum (e.g., absorbance at a specific wavelength in the wavelength range of 1000 to 2700 nm, etc.).

[0195] In addition, data 3 associated with the degree of curing can be obtained from the difference between data 1 and data 2, that is, (data 1)-(data 2). The difference spectrum of the near-infrared absorption spectrum before and after curing obtained in this way, that is, data 3, reflects the changes in the chemical structure and molecular state caused by the curing of the photocurable material, and can therefore be associated with the degree of curing.

[0196] Using the data 1, 2, and 3 acquired in this way, it is possible to predict in advance the adhesion state between the cured product of the photocurable material, that is, the solder resist, and the printed wiring board.

[0197] This application claims the priority based on Japanese Patent Application No. 2022-78942 filed on May 12, 2022. The contents described in the specification and drawings of that application are incorporated herein by reference in their entirety.

[0198] Industrial Applicability

[0199] According to the present invention, it is possible to provide a bonding state prediction system that can be applied to various bonding materials and can predict the bonding state with high accuracy. As a result, in various manufacturing processes, poor bonding can be pre-detected in real time, thereby improving manufacturing efficiency. In particular, the present invention can be applied not only to online defect detection (process management) in the manufacturing process and quality management of the final product, but can also be effectively applied to preliminary research (laboratory research, trial production research, etc.) such as material search, formulation research, and process condition research.

[0200] Explanation of reference numerals: 100 ... prediction system; 110 ... photographing device; 111 ... light source; 112 ... photographing unit; 120 ... processing device; 121 ... data acquisition unit; 122 ... storage unit; 123 ... adhesion state prediction unit; 124 ... image acquisition unit; 125 ... processing unit; 130 ... display unit.

Claims

1. A system for predicting an adhesion state, the system being a system for predicting an adhesion state when an adhesive object is bonded to an adherend, wherein: The bonding state prediction system has: a data acquisition unit that acquires two or more data including physical property value information on two-dimensional coordinates of the object; and The adhesion state prediction unit predicts the adhesion state between the object and the adherend using the two or more acquired data.

2. The adhesion state prediction system according to claim 1, wherein: The data acquisition unit acquires the two-dimensional coordinate information of the object, and acquires the two or more data including the physical property value information in association with the acquired two-dimensional coordinate information.

3. The adhesion state prediction system according to claim 2, wherein: The data acquisition unit comprises: an image acquisition unit that acquires an image representing a state of light reflected or released by the object when irradiated with light; as well as The processing unit acquires the two or more data of the object from the acquired image.

4. The adhesion state prediction system according to claim 3, wherein: The bonding state prediction system further comprises: a light source for irradiating light to the object; and a photographing unit for photographing a light-emitting state of the object that receives light from the light source and emits light, The image acquisition unit acquires the image captured by the imaging unit.

5. The adhesion state prediction system according to claim 1, wherein: The adhesion state prediction unit predicts the adhesion state based on a prediction model generated by machine learning.

6. The adhesion state prediction system according to claim 1, wherein: The adhesion state prediction system further includes a display unit that displays a result predicted by the adhesion state prediction unit.

7. The adhesion state prediction system according to claim 1, wherein: The two or more data are obtained by acquiring data including predetermined physical property value information over time.

8. The adhesion state prediction system according to claim 1, wherein: The two or more data include different types of physical property value information.

9. The adhesion state prediction system according to claim 8, wherein: The two or more data are acquired simultaneously.

10. The adhesion state prediction system according to claim 1, wherein: At least one of the two or more data is spectral characteristic information of the object.

11. The adhesion state prediction system according to claim 10, wherein: The spectral characteristic information is obtained from an image representing a state of light reflected or emitted from the object when the object is irradiated with light of a predetermined wavelength.

12. The adhesion state prediction system according to claim 11, wherein: The image is acquired by a hyperspectral camera.

13. The adhesion state prediction system according to claim 11 or 12, wherein: The object includes a contrast agent whose light emission behavior changes according to the physical property of the object.

14. The adhesion state prediction system according to claim 13, wherein: The contrast agent emits fluorescence when irradiated with predetermined light.

15. The adhesion state prediction system according to claim 10, wherein: The spectral characteristic information is associated with a physical property value selected from the group consisting of elastic modulus, degree of curing, hardness, polarity, and water content.

16. The adhesion state prediction system according to claim 15, wherein: The other of the two or more data includes film thickness information.

17. The adhesion state prediction system according to claim 1, wherein: The object is a thermocompression bonding material, a curable material, or an adhesive material.

18. A method for predicting an adhesive state, the method being a method for predicting an adhesive state when an adhesive object is bonded to an adherend, wherein: The bonding state prediction method has the following features: a step of acquiring two or more data including physical property value information on two-dimensional coordinates of the object; and and predicting a bonding state between the object and the adherend using the two or more acquired data.

19. A program for predicting an adhesion state, the program being for predicting an adhesion state when an adhesive object is bonded to an adherend, wherein: The bonding state prediction program is used to cause a computer to execute: a step of acquiring two or more data including physical property value information on two-dimensional coordinates of the object; and and predicting a bonding state between the object and the adherend using the two or more acquired data.

20. A method for producing an adhesive, wherein: include: A step of predicting the adhesion state between an object having adhesiveness and an adherend by performing the prediction method described in claim 18 on the object having adhesiveness; as well as A step of adjusting processing conditions of the object based on the prediction result.

Images