Laser irradiation device, information processing method, recording medium capable of reading and recording programs, and method for generating learning models.

Abstract

The laser irradiation apparatus includes a laser light source that emits laser light and a control unit that performs control related to irradiating a substrate with laser light. The control unit performs the following steps: acquiring operating parameters including detection values ​​from a detection unit installed in the laser irradiation apparatus; if the operating parameters have already been input, inputting the acquired operating parameters into a learning model for outputting predicted quality information of a product, thereby deriving predicted quality information, wherein the product includes a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the acquired operating parameters.

Description

[Technical Field]

[0001] This invention relates to a laser irradiation device, an information processing method, a recording medium capable of reading and recording programs, and a method for generating learning models. [Previous Technology]

[0002] Laser annealing apparatuses for forming polycrystalline silicon thin films are widely known (e.g., Patent Document 1). The laser annealing apparatus described in Patent Document 1 includes a waveform shaping device for shaping the waveform of a laser light pulse. The amorphous silicon film is irradiated with laser light that has been linearly shaped using this waveform shaping device to form a polycrystalline silicon thin film. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2012-15545 [Summary of the Invention]

[0004] [The problem that the invention aims to solve]

[0005] However, the laser annealing apparatus of Patent Document 1 does not take into account the prediction of the quality information (predicted quality information) of the product manufactured by the laser annealing apparatus based on the operating parameters of the laser annealing apparatus.

[0006] In view of this, the object of the present invention is to provide a laser irradiation apparatus, etc., which predicts quality information (predicts quality information) of the products manufactured by the laser annealing apparatus based on the operating parameters of the laser annealing apparatus. [Means for solving the problem]

[0007] The laser irradiation apparatus of this embodiment includes a laser light source that emits laser light and a control unit that performs related control of irradiating a substrate with laser light. The control unit performs the following: acquiring operating parameters including detection values, the detection values ​​being obtained from a detection unit installed in the laser irradiation apparatus; if operating parameters have been input, inputting the acquired operating parameters into a learning model for outputting predicted quality information of a product, thereby deriving predicted quality information, the product including a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the acquired operating parameters.

[0008] The information processing method of this embodiment performs the following processing in a computer: obtaining operating parameters, which include detection values ​​from the detection unit provided in the laser irradiation device; when the operating parameters are input, inputting the obtained operating parameters into a learning model for outputting predicted quality information of the product, thereby deriving predicted quality information; the product includes a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the obtained operating parameters.

[0009] The program of this embodiment causes the computer to perform the following processing: obtain operating parameters, which include detection values ​​from the detection unit provided in the laser irradiation device; when the operating parameters are input, input the obtained operating parameters into a learning model for outputting predicted quality information of the product, thereby deriving predicted quality information, wherein the product includes a substrate irradiated by laser light; and correlate the derived predicted quality information with the obtained operating parameters and output them.

[0010] The learning model generation method of this embodiment includes: acquiring operating parameters, which include detection values ​​from a detection unit provided in a laser irradiation device; acquiring quality information of a product, wherein the product includes a substrate processed by the laser irradiation device, and the laser irradiation device is controlled using the operating parameters; and generating a learning model for outputting quality information of the product using training data including question data and response data, given the input operating parameters; wherein the question data is composed of the acquired operating parameters, the response data is composed of the acquired quality information, and the product includes a substrate processed by the laser irradiation device. [Effects of the Invention]

[0011] According to the present invention, a laser irradiation device or the like can be provided, which can predict the quality information (predicted quality information) of the products manufactured by the laser annealing device based on the operating parameters of the laser annealing device.

Implementation Method

[0013] [Example 1] The following describes an embodiment of the present invention. Figure 1 shows a system architecture example including the laser annealing apparatus of Example 1. The laser annealing apparatus 1 (laser irradiation apparatus) is, for example, an excimer laser annealing (ELA) apparatus for forming low temperature polycrystalline silicon (LTPS) films.

[0014] The laser annealing apparatus 1 is placed in a manufacturing plant that manufactures semiconductor substrates (substrate 8) such as glass substrates with polycrystalline silicon films already formed. The manufactured substrate 8 is shipped to a final product factory, which manufactures final products including the substrate 8. A product server SS is installed in the final product factory, which stores and manages the quality information of the final product.

[0015] The control device 9 included in the laser annealing apparatus 1 obtains quality information of the final product from the product server SS, for example, via an external network GN such as the Internet. Thus, the laser annealing apparatus 1, including the control device 9, constitutes a quality information acquisition system for obtaining quality information of the final product by means of a plurality of product servers SS that can be communicatively connected via the external network GN. The product servers SS and the laser annealing apparatus 1 are not limited to a particular location; they can also be located at the same location (the final product factory). In this case, the product servers SS and the laser annealing apparatus 1 are connected via the LAN (intranet) of the final product factory.

[0016] Quality information includes the yield, defect frequency, defect location information, and evaluation information of the substrate 8 assembled in the final product. The control device 9 uses the acquired quality information of the final product to generate a learning model 921 as described below, or uses the learning model 921 to perform various processes such as predicting the quality information (predicted quality information) of the final product during the production stage of the substrate 8. The management criteria for the quality information of the final product may differ among multiple final product factories, and the control device 9 may also generate and apply each learning model 921 for these final product factories (the shipping destinations of the substrate 8). Alternatively, information obtained from multiple final product factories that has been normalized, standardized, or averaged may be used to generate a learning model 921 that is universally applicable to these various final product factories.

[0017] Figure 2 shows an example of the architecture of a laser annealing apparatus. Figure 3 shows an example of the architecture of the control device 9 in the laser annealing apparatus. The laser annealing apparatus 1 irradiates a silicon film formed on a substrate 8 with laser light. This allows an amorphous silicon film (a-Si film) to be converted into a polycrystalline silicon film (p-Si film). The substrate 8 is a semiconductor substrate.

[0018] As shown in the figure of this embodiment, in the XYZ three-dimensional orthogonal coordinate system, the Z direction is the vertical direction and is perpendicular to the substrate 8. The XY plane is parallel to the plane in which the silicon film of the substrate 8 is formed. For example, the X direction is the longer direction of the substrate 8, and the Y direction is the shorter direction of the substrate 8. When using a stage 71 that can rotate from 0° to 90° around the Z-axis, the X direction is the shorter direction of the substrate 8, and the Y direction is the longer direction of the substrate 8.

[0019] The laser annealing apparatus 1 includes an annealing optical system 11, a laser irradiation chamber 7, and a control device 9. The laser irradiation chamber 7 houses a base 72 and a stage 71 disposed on the base 72. In the laser annealing apparatus 1, the stage 71 transports the substrate 8 in the +X direction while simultaneously irradiating the silicon film 201 with laser light. In addition, as a detection unit for detecting information related to the emitted laser light, there are a dual-plane phototube 62, an OED sensor 63, a Mura (light source non-uniformity) monitor 64, and an analyzer camera 66.

[0020] The annealing optical system 11 is an optical system for crystallizing the amorphous silicon film formed on the substrate 8, generating laser light for converting it into a polycrystalline silicon film, and irradiating the amorphous silicon film. The annealing optical system 11 includes a laser light source 2, an attenuator 3, a polarization ratio control unit 4, a beam shaping optical system 5, a projection mirror 61, and a projection lens 65, emitting linear laser light.

[0021] The laser light source 2 is a laser generating device that generates pulsed laser light to irradiate the amorphous silicon film (the substrate being processed). The generated laser light is the laser light used to crystallize the amorphous film on the substrate 8 to form a crystalline film, such as a gas laser light with a center wavelength of 308 nm, like an excimer laser. Furthermore, the gas laser light is not limited to excimer laser light; it can also be other laser light such as a Co2 laser.

[0022] In the laser source 2, a gas such as xenon is sealed in the chamber, and two resonator mirrors are arranged opposite each other, sandwiching the gas. One resonator mirror is a total internal reflection mirror that reflects all light, and the other resonator mirror is a partial reflection mirror that transmits a portion of the light. Gas laser light excited by the gas is repeatedly reflected between the resonator mirrors, and the amplified light is emitted from the resonator mirrors as laser light. The laser source 2 repeatedly emits pulsed laser light with a period of, for example, 500Hz to 600Hz. The laser source 2 emits laser light towards the attenuator 3.

[0023] Attenuators 3 attenuate the incident laser light and adjust it to a predetermined energy density. These attenuators have a transmittance characteristic, which indicates the ratio of the emitted laser light to the incident laser light; this transmittance is configured to be variable based on a signal from the control device 9. Attenuators 3 are positioned along the optical path from the laser source 2 to the beam-shaping optical system 5. Attenuators 3 attenuate the laser light emitted from the laser source 2 according to the transmittance.

[0024] The energy density (E) emitted from the attenuator 3 is the value obtained by multiplying the laser light energy density (E0) emitted from the laser light source 2 by the transmittance (T) of the attenuator 3 (E=E0×T). As described later, the control device 9 is configured to specify (derive) and change the transmittance of the attenuator 3 so that the energy density emitted from the attenuator 3 becomes the optimal energy density.

[0025] A polarization ratio control unit 4 is disposed on the emission side of the attenuator 3. The polarization ratio control unit 4 includes, for example, a 1 / 2 wavelength plate (λ / 2 plate) and a polarization beam splitter, which changes the polarization ratio of the P-polarized wave and the S-polarized wave of the incident laser light. That is, the polarization ratio of the laser light emitted from the attenuator 3 is changed by the polarization ratio control unit 4. The polarization ratio control unit 4 is configured to change (variable) the polarization ratio based on the control signal output by the control device 9.

[0026] When the transmittance of the attenuator 3 is changed, the polarization ratio of the laser light emitted from the attenuator 3 changes accordingly. In contrast, the control device 9 is configured to change the polarization ratio of the polarization control unit 4 according to the changed transmittance, thereby controlling the polarization ratio of the laser light emitted from the polarization control unit 4 to be constant.

[0027] When the control device 9 changes the polarization ratio of the polarization ratio control unit 4, it can also refer to information (polarization ratio table) stored in the memory unit 92 of the control device 9 in a table format, for example, to specify (derive) the polarization ratio based on the transmittance. The polarization ratio table defines each polarization ratio corresponding to each transmittance.

[0028] Laser light emitted from the polarization ratio control unit 4 enters the beam shaping optical system 5, which shapes the incoming laser light and generates a laser light beam with a shape suitable for irradiating the silicon film. The beam shaping optical system 5 generates a linear beam along the Y direction.

[0029] The beam shaping optical system 6, for example, splits a single beam into multiple beams (multiple line beams arranged in the Z direction) using a homogenizer composed of a lens array. After being split into multiple beams, they can be combined using a condenser lens to create a line beam. The beam shaping optical system 6 then emits the generated (shaped) line laser light into the incident mirror 61.

[0030] The reflector 61 is a rectangular reflector extending in the Y direction. The beam shaping optical system 6 reflects the laser light into multiple linear beams that have already been generated. The reflector 61 is, for example, a dichroic filter, and is a partial reflector that transmits a portion of the light. The reflector 61 reflects the linear laser light and generates reflected light, while simultaneously allowing a portion of the linear laser light to pass through and generating transmitted light. The reflector 61 illuminates the silicon film of the substrate 8 with the reflected laser light, and the transmitted light emanates the laser light to a pulse detector such as a dual-plane phototube.

[0031] A projection lens 65 is disposed above the substrate 8. The projection lens 65 includes a plurality of lenses for projecting laser light onto the substrate 8, i.e., the silicon film. The projection lens 65 focuses the laser light onto the substrate 8, forming a linear irradiation area along the Y direction on the substrate 8. That is, the laser light on the substrate 8 is a line beam with the Y direction as the longer direction. Furthermore, the substrate 8 is moved in the +X direction while the silicon film is irradiated with laser light. In this way, laser light can be irradiated on a strip-shaped area where the length of the irradiation area in the Y direction is equal to the width.

[0032] The linear laser beam that illuminates the incident mirror 61 becomes a beam shape with an expanded minor axis width, that is, after exiting the condenser lens, the minor axis width becomes a shape that is somewhat expanded and collapsed. The laser beam reflected by the incident mirror 61 passes through the projection lens 65, thereby shaping the laser beam into a linear laser beam with a minor axis width of about 1 / 5.

[0033] The dual-plane phototube 62 is adjacent to the beam shaping optical system 6 and is disposed at the end of the annealing optical system 11. Based on the transmitted light through the transmission mirror 61, it detects the pulse waveform of the laser light emitted from the laser source 2. The dual-plane phototube 62 outputs (transmits) the detected pulse waveform to the control device 9.

[0034] The OED sensor 63 includes a light sensor that detects reflected light (reflected light reflected by the substrate 8) emitted from a light source other than the laser light source 2, and obtains information related to the crystal surface on the substrate 8. The OED sensor 63 outputs the brightness (detection value) of the detected reflected light (as a signal transmission) to the control device 9.

[0035] The Mura monitor 64 includes a line camera that captures an area of ​​interest on the substrate 8 irradiated by laser light, detects the average brightness of the area of ​​interest contained in the captured image, and obtains information related to the scattered light of the surface shape of the substrate 8. The Mura monitor 64 outputs the detected average brightness (detection value) of the substrate 8 (area of ​​interest) to the control device 9 (as a signal transmission).

[0036] The analyzer camera 66 is a sensor (line beam sensor), such as a beam analyzer, that detects information related to the shape of the laser light shaped into a line beam by the projection lens 65. The analyzer camera 66 is, for example, disposed on the side of the stage 71, and is configured such that the top of the analyzer camera 66 and the substrate 8 on which the stage 71 is placed are at the same height. The linearly shaped laser light from the annealing optical system 11 illuminates the top of the analyzer camera 66. The analyzer camera 66 includes, for example, an imaging unit such as a CMOS camera, which captures the linearly shaped laser light to obtain images (captured images) and other information (data) related to the shape of the laser light. The analyzer camera 66 can also detect, for example, information related to the minor and major axis shapes of the rectangular beam, the skewing or concavity of the axis, the tilt when viewing the line beam stereoscopically, and the angle or curvature between adjacent surfaces, as information related to the shape of the linearly shaped laser light. The analyzer camera 66 can further detect information related to the shape of the raw (initial) beam before line beam shaping. In addition to the analyzer camera 66 in this embodiment, a line beam sensor for obtaining laser beam shape-related information can also be placed near, for example, a dual-plane phototube 62, such that the dual-plane phototube 62 is in a different direction from the Y-axis.

[0037] The control device 9 is an information processing device such as a personal computer or server device, which performs overall or integrated control or management of the laser annealing apparatus 1. The control device 9 includes a control unit 91, a memory unit 92, a communication unit 93, and an input-output I / O 94. Through the communication unit 93 or the input-output I / O 94, it can communicate with the control devices (other control devices) that control each optical system in the laser light source 2 or the annealing optical system 11. The control device 9 can communicate with various measuring devices such as pulse counters and photodetectors included in the laser annealing apparatus 1, and perform various controls on the laser light source 2 or the annealing optical system 11 based on the measuring data output by these various measuring devices.

[0038] The control unit 91 includes one or more CPUs (Central Processing Units), MPUs (Micro-Processing Units), GPUs (Graphics Processing Units), etc., and computing processing devices with timing functions. By reading out the program P (program product) stored in the memory unit 92, it performs various information processing and control processing of the various optical systems included in the laser light source 2 or the annealed optical system 11.

[0039] The memory unit 92 includes volatile memory fields such as SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), and flash memory, as well as non-volatile memory fields such as EEPROM or hard disks. The memory unit 92 pre-stores the program P (program product) and the data to be referenced during processing. The program P stored in the memory unit 92 can also be a program P (program product) read from the memory medium 920 readable by the control unit 91. Alternatively, the program P (program product) can be downloaded from an external computer (not shown) connected to a communication network (not shown) and stored in the memory unit 92. The memory unit 92 stores the actual file of the learning model 921 as described later. The actual file of the learning model 921 can also be configured as a module included in the program P (program product).

[0040] The communication unit 93 is, for example, a communication module or communication interface based on the ETHERNET (registered trademark) standard, and the communication unit 93 is connected to an Ethernet cable. The communication unit 93 is not limited to wired cases such as the Ethernet cable, and may also be a communication interface corresponding to narrow-area wireless communication modules such as Wi-Fi (registered trademark), Bluetooth (registered trademark), or wide-area wireless communication modules such as 4G, 5G. The control device 9 may be a product server SS that is connected to, for example, an external network GN through the communication unit 93.

[0041] The input-output I / F 94 is, for example, a communication interface based on communication standards such as RS232C or USB. The input-output I / F 94 is connected to an input device such as a keyboard, or a display device 941 such as an LCD. The control device 9 can also obtain various detection values ​​from the detection unit of the dual-plane phototube 62, the OED sensor 63, the Mura monitor 64, or the analyzer camera 66 through the input-output I / F 94.

[0042] Figure 4 shows an explanatory diagram of an example of the learning model 921. The control unit 91 of the control device 9 uses training data to train a neural network to learn, and generates a learning model 921 when the operating parameters of the laser device 1 are input. The learning model 921 outputs quality information (predicted quality information) of the product of the substrate 8 irradiated by laser light. The predicted quality information is the quality information predicted (estimated) by the learning model 921.

[0043] The operating parameters include detection values ​​(parameters) from detection units such as the OED sensor 63 installed in the laser annealing apparatus 1. These detection values ​​are measured at a predetermined period, and the operating parameters may be the standard deviation or average of multiple detection values ​​detected at that predetermined period. The product including the substrate 8 irradiated by laser light is, for example, a mobile terminal such as a liquid crystal display or a smartphone. The quality information of the final product includes things related to defects detected in the substrate 8 when the substrate 8 is assembled into the final product, including, for example, yield, defect frequency, or the location of the defect. Furthermore, the quality information may also include qualitative information such as evaluation information given by quality managers at the final product factory.

[0044] In addition to the detection values ​​from the OED sensor 63 and other detection units, the operating parameters may also include parameters related to the state of the laser annealing apparatus 1 (state parameters) and parameters related to the control of the laser annealing apparatus 1 (control parameters). State parameters include, for example, parameters related to the state of the laser light source 2 and parameters related to the state of the laser irradiation chamber 7 (processing chamber) on the substrate 8. Control parameters include, for example, parameters related to the control of the laser light source 2, parameters related to the control of the optical system (annealing optical system 11) used to shape the laser light emitted from the laser light source 2, and parameters related to the control of the laser irradiation chamber 7 (processing chamber) on the substrate 8.

[0045] The various parameters included in the operating parameters are not limited to those described above. For example, all the data contained in the management screen of the laser annealing apparatus 1 (Figure 7) described later can be included in the operating parameters. The control unit 91 of the control device 9 obtains detection values ​​from various sensors such as the OED sensor 63, temperature sensors, vibration sensors, pressure sensors, and cameras installed at various locations in the laser annealing apparatus 1, and references the operating records (log data) stored in the memory unit 92 to obtain the operating parameters.

[0046] The training data includes question data and answer data, which are interconnected and stored in the memory unit 92 of the control device 9; the question data includes detection values, status parameters, and operating parameters including control parameters from detection units such as the OED sensor 63; the answer data consists of product quality information including yield. The original data for the question data that forms this training data can be generated by collecting, for example, operational experience data from multiple laser annealing devices 1.

[0047] The original data used as training data can be obtained, as described above, from a product server SS that stores and manages quality information of the final product via, for example, an external network GN. This final product is assembled with a substrate 8 processed by the laser annealing apparatus 1. Furthermore, the control device 9 of the laser annealing apparatus 1 can obtain quality information by referring to a memory medium storing the quality information of the final product.

[0048] A neural network (learning model 921) learned using training data is assumed to be a program module as part of artificial intelligence software. The learning model 921 is used in the control device 9, which is implemented by the control device 9 having computational processing capabilities to form a neural network system.

[0049] The learning model 921 includes a DNN (Deep Neural Network), including an input layer for receiving inputs containing operating parameters of the detection values, an intermediate layer for extracting the feature values ​​of the operating parameters, and an output layer set to output quality information (prediction quality information).

[0050] The input layer has multiple neurons that receive inputs including operational parameters such as detection values ​​and deliver the input values ​​to the intermediate layer. The intermediate layer, defined using activation functions such as the ReLU function or the sigmoid function, has multiple neurons that extract feature values ​​from each input value and deliver the extracted feature values ​​to the output layer. The weighting coefficients and bias values ​​of these activation parameters are optimized using the backpropagation method. The output layer, for example, is composed of a fully associative layer and, based on the feature values ​​output by the intermediate layer, outputs quality information (predicted quality information) including yield.

[0051] In this embodiment, the learning model 921 is set as DNN, but it is not limited to this. It can also be a learning model 921 constructed by other learning algorithms such as neural networks other than DNN, Transformer, RNN (Recurrent Neural Network), LSTM (Long-short term model), CNN, SVM (Support Vector Macgine), Bayesian network, linear regression, regression tree, multiple linear regression, random forest, ensemble, etc.

[0052] The control device 9 in the laser annealing apparatus 1 generates a learning model 921, but it is not limited to this. The learning model 921 can also be learned and generated using an external server device such as a cloud server other than the control device 9. Although the learning model 921 is used in the control device 9, it is not limited to this. The control device 9 can also communicate with a cloud server, such as one connected to the Internet, through the communication unit 93 to obtain the predicted quality information (yield, etc.) output by the learning model 921 installed on the cloud server.

[0053] Figure 5 shows an example flow of the processing sequence of the control unit 91 (when the learning model 921 is learning). The control unit 91 of the control device 9 included in the laser annealing apparatus 1 receives operations from an operator, such as those connected to an input keyboard, and performs the following processing based on the received operations.

[0054] The control unit 91 of the control device 9 acquires operating parameters (S11). The control unit 91 of the control device 9 acquires the detection values ​​of various sensors such as the OED sensor 63, temperature sensors, vibration sensors, pressure sensors and cameras installed at various locations of the laser annealing apparatus 1, and obtains operating parameters containing this data by referring to the operating log data stored in the memory unit 92.

[0055] The control unit 91 of the control device 9 obtains the quality information of the final product (S12). The control unit 91 of the control device 9 can obtain the quality information of the final product from the product server SS, which stores and manages the quality information of the final product, and the final product is assembled with a substrate 8 processed by the laser annealing apparatus 1. Alternatively, the control device 9 of the laser annealing apparatus 1 can obtain the quality information by referring to a memory medium storing the quality information of the final product.

[0056] The control unit 91 of the control device 9 generates training data using the acquired operating parameters and quality information of the final product (S13). The control unit 91 of the control device 9 generates training data in which the operating parameters are set as problem data and the quality information is set as response data. When generating the training data, the control unit 91 of the control device 9 can perform standard deviation processing, averaging processing, standardization processing, or dimension reduction processing based on the detection values ​​at multiple time points.

[0057] The control unit 91 of the control device 9 uses the generated training data to generate a learning model 921 (S14). The control unit 91 of the control device 9 uses the generated training data to generate a learning model 921 by learning, for example, a neural network.

[0058] When there are multiple final product factories to which the substrate 8 is shipped, the control unit 91 of the control device 9 can generate different learning models 921 for each of these final product factories. Alternatively, the control unit 91 of the control device 9 can generate different learning models 921 based on the allocation or category of the final products assembled with the substrate 8. Alternatively, different learning models 921 can be generated based on combinations of final product factories and the allocation of final products.

[0059] When the management standards for quality information obtained from the product servers SS of each final product factory are different, or when the quality information includes qualitative evaluation information, the control unit 91 of the control device 9 can normalize, standardize, or average the quality information obtained and collected from these final product factories. The control unit 91 of the control device 9 can also use the normalized quality information to generate a learning model 921 that is applicable to all these final product factories.

[0060] Figure 6 shows an example flow of the processing sequence of the control unit 91 (when the learning model 921 is running). The control unit 91 of the control device 9 included in the laser annealing apparatus 1 receives operations from an operator, such as those connected to an input keyboard, and performs the following processing based on the received operations.

[0061] The control unit 91 of the control device 9 acquires operating parameters (S101). The control unit 91 of the control device 9 acquires the detection values ​​of various sensors such as the OED sensor 63, temperature sensors, vibration sensors, pressure sensors and cameras installed at various locations of the laser annealing apparatus 1, and acquires (generates) operating parameters containing this data by referring to the operating log data stored in the memory unit 92.

[0062] The control unit 91 of the control device 9 inputs the acquired operating parameters into the learning model 921 and obtains predicted quality information of the final product (S102). The control unit 91 of the control device 9 inputs the acquired operating parameters into the learning model 921. The learning model 921 outputs (estimates) predicted quality information of the final product, such as yield, based on the input operating parameters. By obtaining the predicted quality information (yield, etc.) output by the learning model 921, the control unit 91 of the control device 9 can deduce the predicted quality information.

[0063] The control unit 91 of the control device 9 outputs the predicted quality information of the final product obtained from the learning model 921 (S103). The control unit 91 of the control device 9 associates the predicted quality information of the final product obtained from the learning model 921 with the operating parameters, and outputs it to, for example, a display device 941, to notify the manager of the laser annealing apparatus 1 of the predicted quality information of the final product, such as the yield estimated from the operating parameters at the current point in time.

[0064] The control unit 91 of the control device 9 determines whether the yield included in the predicted quality information is less than a preset threshold (S104). The threshold relative to the yield included in the predicted quality information is stored in, for example, the memory unit 92 of the control device 9. The control unit 91 of the control device 9 refers to the threshold stored in the memory unit 92 to determine whether the yield included in the predicted quality information is less than the threshold.

[0065] When the yield is not less than the threshold (S104: No), that is, when the yield included in the predicted quality information is above the threshold, the control unit 91 of the control device 9 should perform the processing of S101 again and perform loop processing. When the yield is not less than the threshold, that is, above the threshold, the control unit 91 of the control device 9 determines that the operating parameters at the current point of time are appropriate, and should perform the processing of S101 and perform loop processing.

[0066] When the yield is less than the threshold (S104: Yes), the control unit 91 of the control device 9 outputs a notification signal (S105) to indicate that the yield is less than the threshold. When the yield is less than the threshold, the control unit 91 of the control device 9 determines that the operating parameters at the current point of operation are inappropriate, and outputs a notification signal indicating that the yield is less than the threshold to, for example, a display device or a mobile terminal of the manager of the laser annealing device 1.

[0067] After the control unit 91 of the control device 9 performs the processing of S105, it should perform the processing of S101 again. By performing the loop processing, the suitability of the operating parameters based on the quality information of the final product can be continuously monitored.

[0068] Figure 7 illustrates an example of the management screen of the laser annealing apparatus 1. The control unit 91 of the control device 9 uses the acquired operating parameters and the derived (derived) predicted quality information to generate a management screen (screen data) as an example shown in this embodiment, and outputs it to a display device 941, etc., for example.

[0069] The management screen of the laser annealing apparatus 1 includes an area that displays laser-related data, optical system-related data, processing chamber-related data, and substrate observation-related data in a list format, as well as an area that displays estimated quality information.

[0070] The display area for laser-related data includes the display allocation for the laser output system, control system, laser gas system, maintenance system, and utility system. The laser output system display allocation shows the laser pulse energy, the standard deviation (σ) of the laser pulse energy, and the pulse waveform. The control system display allocation shows the electrode voltage, oscillation frequency, and resonator temperature. The laser gas system display allocation shows the gas ratio and pressure. The maintenance system display allocation shows the exchange status and conditions of consumables. The utility system display allocation shows the cooler cooling temperature, flow rate, and power supply voltage.

[0071] The display area for optical system-related data includes the display allocation of the line beam minor axis shape, line beam major axis shape, and raw beam shape, as well as the display items for transmittance and polarization ratio. The display allocation for the line beam minor axis shape shows the minor axis width, shoulder width, standard deviation (σ) within the minor axis width, and tilt. The display allocation for the line beam major axis shape shows the major axis width and standard deviation (σ) within the major axis width. The display allocation for the raw beam shape shows the shape, position, emission angle, and intensity. The transmittance display shows the transmittance of the attenuator 3. The polarization ratio display shows the polarization ratio of the polarization ratio control unit 4.

[0072] The display area for processing chamber-related data includes items such as processing speed, irradiation environment, stage surface flatness, and processing chamber vibration. The processing speed display shows the stage speed and speed stability (fluctuation). The irradiation environment display shows the oxygen concentration, distribution, and nitrogen (N2) flow rate. The stage surface flatness display shows the displacement sensor value. The processing chamber vibration display shows floor vibration and vibration within the stage. The display area for substrate observation-related data includes the detection values ​​from the Mura monitor 64 and the OED sensor 63.

[0073] The area displaying the predicted quality information includes a graphical display area for showing the yield over time shift and a list display area for displaying the predicted preset quality information in a list format. The horizontal axis of the graph showing the yield over time shift represents the elapsed time, and the vertical axis represents the yield. A threshold value is preset in the yield. The area displaying the list of predicted quality information includes display items for yield, frequency of defects, and location information of defects on the substrate 8.

[0074] In this way, the operating parameters obtained from the operation of the laser annealing apparatus 1 and the yield (predicted quality information of the final product assembled with the substrate 8) derived (estimated) based on the operating parameters are displayed on the screen in relation to each other, thereby improving the visibility for the managers of the laser annealing apparatus 1.

[0075] According to this embodiment, a learning model 921 is used to acquire and output predicted quality information (including predicted quality information of the final product of the substrate 8); this predicted quality information is derived (predicted) based on operating parameters including detection values ​​obtained from the detector. Therefore, from the viewpoint of the quality information of the product of the substrate 8 irradiated by laser light, the suitability of the detection values ​​(operating parameters) can be determined (state diagnosis), and the administrator of the laser annealing apparatus 1 (laser irradiation apparatus) can be notified. That is, based on the operating parameters including the detection values, the quality information (predicted quality information) of the product (final product) of the substrate 8 manufactured by the laser annealing apparatus 1 (laser irradiation apparatus) is predicted, thereby performing the correlation between the operating parameters and the predicted quality information of the product, and seeking to adjust the operating parameters based on this correlation, thus enabling efficient control related to laser light.

[0076] According to this embodiment, when the yield of the product contained in the predicted quality information derived by the learning model 921 is less than a preset threshold, a notification signal showing this meaning is output, which can promote alertness to the manager of the laser annealing device 1 and others regarding the judgment of whether the laser annealing device 1 should continue to operate.

[0077] According to this embodiment, since the detection unit includes various detection units such as the OED sensor 63, the Mura monitor 64, the dual-plane phototube 62, and the analyzer camera 66, the multiple detection values ​​detected by these detection units can be used as input data for the learning model 921, thereby improving the prediction accuracy of the learning model 921. The operating parameters input to the learning model 921 include the standard deviation calculated based on the multiple detection values ​​detected in a predetermined period, which further improves the prediction accuracy of the learning model 921.

[0078] According to this embodiment, the operating parameters input to the learning model 921 include parameters related to the state of the laser light source 2 and parameters related to the state of the laser irradiation chamber 7 (processing chamber) used to mount the substrate 8, thereby improving the prediction accuracy of the learning model 921.

[0079] According to this embodiment, the operating parameters input to the learning model 921 include parameters related to the control of the laser light source 2, parameters related to the control of the optical system, and parameters related to the control of the laser irradiation chamber 7 (processing chamber), thereby improving the prediction accuracy of the learning model 921.

[0080] [Example 2] Figure 8 shows an example flow chart of the processing sequence (deriving operating parameters) of the control unit 91 in Example 2. The control unit 91 of the control device 9 included in the electro-annealing apparatus 1 accepts operations from the operator via, for example, a keyboard connected to input / output, and performs the following processing based on the accepted operations.

[0081] The control unit 91 of the control device 9 acquires operating parameters (S201). The control unit 91 of the control device 9 inputs the acquired operating parameters into the learning model 921 and acquires the predicted quality information of the final product (S202). The control unit 91 of the control device 9 outputs the predicted quality information of the final product acquired from the learning model 921 (S203). The control unit 91 of the control device 9 determines whether the yield contained in the predicted quality information is less than a preset threshold value (S204). The control unit 91 of the control device 9 outputs a notification signal (S205), displaying the key point that the yield is less than the threshold value. The control unit 91 of the control device 9 performs the processing of S201 to S205 as in the implementation of processing S101 to S105.

[0082] After the processing in S205, the control unit 91 of the control device 9 generates a plurality of candidate operating parameters when changing the operating parameters (S206). The control unit 91 of the control device 9 changes the values ​​contained in the current operating parameters successively (stages) within a predetermined range to generate a plurality of operating parameters as candidate operating parameters (candidate parameters). When generating the plurality of candidate parameters, the control unit 91 of the control device 9 can also use the current operating parameters as a reference to successively (stages) change the control parameters of the laser light source 2, such as electrode voltage or oscillation frequency, the control parameters of the annealing optical system 11, such as transmittance or polarization ratio, or the control parameters related to the laser irradiation chamber 7, such as processing speed, and combine these successively changing parameters.

[0083] The control unit 91 of the control device 9 inputs complex candidate parameters into the learning model 921 and obtains complex prediction quality information (S207). By repeatedly inputting each generated complex candidate parameter into the learning model 921, the control unit 91 of the control device 9 can obtain complex prediction quality information corresponding to each candidate parameter.

[0084] The control unit 91 of the control device 9 sets the highest predicted quality information among the acquired complex predicted quality information as the target quality information (S208). The control unit 91 of the control device 9 sets, for example, the predicted quality information with the highest yield as the target quality information among each predicted quality information inferred based on these candidate parameters. Needless to say, the set target quality information (target yield) will be higher than the predicted product information (yield) output (inferred) by the learning model 921 based on the current number of operations.

[0085] The control unit 91 of the control device 9 specifies the operating parameters corresponding to the set target quality information (S209). The control unit 91 of the control device 9 specifies the operating parameters, which are the input data when the learning model 921 outputs (predicts) the target quality information, thereby deriving the operating parameters corresponding to the target quality information.

[0086] The control unit 91 of the control device 9 restarts operation using the specified operating parameters (S210). When the laser annealing apparatus 1 replaces the substrate 8 or the box containing multiple substrates 8 is replaced, the control unit 91 of the control device 9 changes the irradiation conditions from the current operating parameters to the operating parameters specified in S209, thereby restarting the irradiation of the substrate 8 with laser light.

[0087] According to this embodiment, the control unit 91 of the control device 9 sets target quality information of a higher quality than the predicted quality information predicted by the learning model 921, based on the current operating parameters. The control unit 91 of the control device 9 changes the values ​​contained in the current operating parameters successively (in stages) within a predetermined range to generate a plurality of operating parameters as candidate operating parameters. The control unit 91 of the control device 9 inputs each generated plurality of candidate parameters into the learning model 921, thereby obtaining the predicted quality information corresponding to each candidate parameter.

[0088] The control unit 91 of the control device 9 sets, for example, the predicted quality information with the highest yield as the target quality information among each predicted quality information inferred based on these candidate parameters, and specifies operating parameters, which are the input data when the learning model 921 outputs (predicts) the target quality information. The control unit 91 of the control device 9 performs control related to the irradiation of laser light onto the substrate 8 based on the operating parameters corresponding to the target quality information, thereby improving the yield of the final product of the substrate 8 assembled with the laser annealing apparatus 1 and improving the quality information of the final product.

[0089] [Other Embodiments] Figures 9, 10, 11, 12, and 13 show cross-sectional views of the steps of a semiconductor device manufacturing method related to another embodiment (method of manufacturing a semiconductor device). As another embodiment, a semiconductor manufacturing method using the laser annealing apparatus 1 related to the foregoing embodiment will be described. In the following semiconductor device manufacturing method, the step of crystallizing an amorphous semiconductor film is to perform an annealing process, which is performed using the laser annealing apparatus 1 of Embodiments 1 to 2.

[0090] The semiconductor device is a semiconductor device having a TFT (thin-film transistor), where an amorphous silicon film 84 is crystallized by irradiation with laser light to form a polycrystalline silicon film 85. The polycrystalline silicon film 85 is used as a semiconductor layer having a TFT source region, a channel region, and a drain region.

[0091] The laser annealing apparatus 1 described above is suitable for manufacturing TFT array substrates. The following describes a manufacturing method related to a semiconductor device having TFTs.

[0092] First, as shown in Figure 9, a gate electrode 82 is formed on a glass substrate 81 (substrate 8). The gate electrode 82 can be, for example, a thin metal film containing aluminum. Next, as shown in Figure 10, a gate insulating film 83 is formed on the gate electrode 82. The gate insulating film 83 is formed in a manner that covers the gate electrode 82. Then, as shown in Figure 11, an amorphous silicon film 84 is formed on the gate insulating film 83. The amorphous silicon film 84 is disposed overlapping the gate electrode 82 through the gate insulating film 83.

[0093] The gate insulating film 83 is a silicon nitride film (SiNx film), a silicon oxide film (SiO2 film), or a laminate of such films. Specifically, the gate insulating film 83 and the amorphous silicon film 84 are continuously deposited using CVD (chemical vapor deposition). The glass substrate 81 on which the amorphous silicon film 84 is attached becomes a semiconductor film in the laser annealing apparatus 1 (laser irradiation apparatus).

[0094] Next, as shown in Figure 12, the amorphous silicon film 84 is irradiated with laser light L3 using the laser annealing apparatus 1 described above, causing the amorphous silicon film 84 to crystallize and form a polycrystalline silicon film 85. Thereby, the silicon-crystallized polycrystalline silicon film 85 is formed on the gate insulating film 83.

[0095] Subsequently, as shown in Figure 13, an interlayer insulating film 86, a source electrode 87a, and a drain electrode 87b are formed on the polycrystalline silicon film 85. The interlayer insulating film 86, the source electrode 87a, and the drain electrode 87b can be formed using conventional photolithography or film deposition methods. Subsequent manufacturing steps will vary depending on the final manufactured apparatus, and therefore their description is omitted.

[0096] By using the semiconductor device manufacturing method described above, a semiconductor device having a TFT comprising a polycrystalline semiconductor film can be manufactured. This semiconductor device is suitable for control of high-precision displays such as organic EL (electroluminescent) displays. As described above, by suppressing light source / brightness unevenness (mura) of the polycrystalline silicon film 85, display devices with superior display characteristics can be manufactured with high productivity.

[0097] During a series of processing steps, the control device 9 of the laser annealing apparatus 1 derives predicted quality information of the final product including the substrate 8 based on the obtained operating parameters, and outputs the predicted quality information to the display device 941. By performing the correlation between the operating parameters and the predicted quality information of the product, monitoring related to the correctness of the calculation parameters from the perspective of the quality information of the final product can be achieved, and the correctness of the operating parameters can be supported, and the control related to the laser annealing apparatus 1 can be executed efficiently.

[0098] Furthermore, this disclosure is not limited to the above embodiments, and appropriate modifications can be made without departing from the essential points. For example, instead of being limited to forming a polycrystalline silicon film 85 by irradiating an amorphous silicon film 84 with laser light, a microcrystalline silicon film can also be formed by irradiating an amorphous silicon film 84 with laser light. Also, a crystallized film can be formed by irradiating an amorphous film other than a silicon film with laser light.

[0099] The embodiments disclosed herein are illustrative of all key points and should not be considered as limiting the invention. The technical features described in the various embodiments can be combined with each other, and the scope of the invention is intended to include all modifications within the scope of the claims and scopes equivalent to the scope of the claims. [Simplified Explanation of the Diagram]

[0012] Figure 1 shows a system architecture example including the laser annealing apparatus of Embodiment 1. Figure 2 shows an architecture example of a laser annealing apparatus. Figure 3 shows an architecture example of a control device in a laser annealing apparatus. Figure 4 shows an explanation of an example of a learning model. Figure 5 shows an example flow chart of the processing sequence of the control unit (during learning model learning). Figure 6 shows an example flow chart of the processing sequence of the control unit (during learning model operation). Figure 7 illustrates an example of a management screen for a laser annealing apparatus. Figure 8 shows an example flow chart of the processing sequence of the control unit of Embodiment 2 (exporting operating parameters). Figure 9 shows a cross-sectional view of the steps of a semiconductor device manufacturing method related to another embodiment (semiconductor device manufacturing method). Figure 10 shows a cross-sectional view of the steps of a semiconductor device manufacturing method related to another embodiment (semiconductor device manufacturing method). Figure 11 shows a cross-sectional view of the steps of a semiconductor device manufacturing method related to another embodiment (semiconductor device manufacturing method). Figure 12 shows a cross-sectional view of the steps of a semiconductor device manufacturing method related to another embodiment (semiconductor device manufacturing method). Figure 13 shows a cross-sectional view of the steps of a semiconductor device manufacturing method related to other embodiments (method of manufacturing a semiconductor device).

Claims

1. A laser irradiation device, comprising: A laser light source that emits laser light, and a control unit that performs related control of irradiating a substrate with laser light; the control unit performs the following: acquiring operating parameters including detection values, the detection values ​​being obtained from a detection unit installed in the laser irradiation device; the operating parameters including control parameters for controlling the laser irradiation device; if the operating parameters have been input, inputting the acquired operating parameters into a learning model for outputting predicted quality information of a product, thereby deriving predicted quality information, the product including a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the acquired operating parameters.

2. The laser irradiation apparatus of claim 1, wherein the predicted quality information derived from the aforementioned learning model includes information related to the yield of the aforementioned product; the aforementioned control unit outputs a notification signal indicating that the yield is less than a preset threshold when the yield of the aforementioned product is less than a preset threshold.

3. The laser irradiation apparatus of claim 1, wherein the aforementioned detection unit includes at least one of an OED sensor, a Mura monitor, a dual-plane phototube, and an analyzer camera; the aforementioned OED sensor detects information related to the crystalline surface on the aforementioned substrate; the aforementioned Mura monitor detects information related to the scattered light of the surface shape of the aforementioned substrate; the aforementioned dual-plane phototube detects the pulse waveform of the laser light emitted from the aforementioned laser light source; and the aforementioned analyzer camera detects information related to the shape of the laser light shaped into a line beam.

4. The laser irradiation apparatus of claim 1, wherein the aforementioned detection unit acquires multiple detection values ​​at a predetermined period; the aforementioned operating parameters include a standard deviation calculated based on the multiple detection values.

5. The laser irradiation apparatus of claim 1, wherein the aforementioned operating parameters include at least one of parameters related to the state of the aforementioned laser light source and parameters related to the state of the laser irradiation chamber on which the aforementioned substrate is mounted.

6. The laser irradiation apparatus of claim 1, wherein the aforementioned control parameters include at least one of the parameters related to the control of the aforementioned laser light source, the parameters related to the control of the optical system for shaping the laser light emitted from the aforementioned laser light source, and the parameters related to the control of the laser irradiation chamber on which the aforementioned substrate is mounted.

7. The laser irradiation apparatus of claim 6, wherein the aforementioned control unit performs the following: setting target quality information that is of higher quality than the derived predicted quality information; deriving control parameters corresponding to the set target quality information; and performing related control of laser light irradiating the aforementioned substrate based on the derived control parameters.

8. An information processing method, comprising performing the following processing in a computer: acquiring operating parameters, including detection values ​​from a detection unit provided in a laser irradiation apparatus; the operating parameters including control parameters for controlling the laser irradiation apparatus; when the operating parameters are input, inputting the acquired operating parameters into a learning model for outputting predicted quality information of a product, thereby deriving predicted quality information; the product including a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the acquired operating parameters.

9. A recording medium capable of reading and recording programs, wherein the recorded programs cause a computer to perform the following processing: acquiring operating parameters, including detection values ​​from a detection unit disposed in a laser irradiation apparatus; the operating parameters including control parameters for controlling the laser irradiation apparatus; when operating parameters are input, inputting the acquired operating parameters into a learning model for outputting predicted quality information of a product, thereby deriving predicted quality information; the product comprising a substrate irradiated by laser light; and correlating and outputting the derived predicted quality information and the acquired operating parameters.

10. A method for generating a learning model, comprising: The system acquires operating parameters, including detection values ​​from a detection unit installed in the laser irradiation apparatus; these operating parameters include control parameters for controlling the laser irradiation apparatus. It acquires quality information about a product, including a substrate processed by the laser irradiation apparatus, which is controlled using these operating parameters. Using training data containing question data and response data, it generates a learning model for outputting product quality information, given the input operating parameters; the question data is composed of the acquired operating parameters, and the response data is composed of the acquired quality information; the product includes a substrate processed by the laser irradiation apparatus.

Images