Computer program, information processing method, and information processing device

Abstract

Provided are a computer program, an information processing method, and an information processing device.　The computer program, which is for outputting information of a processing parameter to be applied to a processing step for correcting deformation of a substrate, causes a computer to execute processing for receiving a constraint condition in the processing step, predicting an amount of deformation of the substrate in accordance with the given processing parameter, evaluating the prediction result of the amount of deformation, searching for the processing parameter to be applied to the processing step under the received constraint condition, and outputting information of a candidate for the processing parameter obtained as the search result.

Description

Computer Program, Information Processing Method, and Information Processing Apparatus

[0001] The present disclosure relates to a computer program, an information processing method, and an information processing apparatus.

[0002] In the process of substrate processing, deformation such as warping may occur in the substrate being processed. As a technique for correcting such deformation, Patent Document 1 discloses a technique of generating a mask pattern for partially exposing a stress film and exposing the stress film using the generated mask pattern to change the stress applied to the substrate in the in-plane direction.

[0003] Japanese Patent Application Laid-Open No. 2022-000917

[0004] The present disclosure provides a computer program, an information processing method, and an information processing apparatus that can present processing parameters to be applied to a processing step for correcting substrate deformation.

[0005] The computer program of the present disclosure is a computer program that outputs information on processing parameters to be applied to a processing step for correcting substrate deformation, receives constraint conditions in the processing step, predicts the amount of deformation of the substrate according to the given processing parameters, evaluates the prediction result of the amount of deformation, searches for processing parameters to be applied to the processing step under the received constraint conditions, and outputs information on candidates for the processing parameters obtained as a search result to cause a computer to execute the processing.

[0006] According to the present disclosure, it is possible to present processing parameters to be applied to a processing step for correcting substrate deformation.

[0007] This is an explanatory diagram illustrating the outline of the experimental condition generation system according to Embodiment 1. This is a block diagram showing the internal configuration of the information processing device. This is an explanatory diagram illustrating the outline of the processing performed by the information processing device. This is a flowchart showing the experimental condition search procedure performed by the information processing device. This is a schematic diagram showing an example of output for candidate processing parameters. This is a schematic diagram showing a modified version of the prediction model. This is a schematic diagram showing an example of an input screen for accepting constraints on processing parameters. This is a schematic diagram showing an example of a modification screen for accepting modifications to processing parameters. This is a schematic diagram showing an example of displaying processing parameters extracted from a database. This is a schematic diagram showing a first example of a mask pattern in Embodiment 5. This is a schematic diagram showing a second example of a mask pattern in Embodiment 5. This is a schematic diagram showing a third example of a mask pattern in Embodiment 5. This is a schematic diagram showing a fourth example of a mask pattern in Embodiment 5. This is a schematic diagram showing a fifth example of a mask pattern in Embodiment 5.

[0008] An embodiment will be described below with reference to the drawings. In this description, the same elements or elements having the same function will be denoted by the same reference numeral, and redundant descriptions will be omitted.

[0009] (Embodiment 1) Figure 1 is an explanatory diagram illustrating the outline of the experimental condition generation system according to Embodiment 1. The experimental condition generation system according to Embodiment 1 comprises a substrate processing device 100 that performs a processing step to correct the deformation of the substrate, and an information processing device 200 that outputs information on processing parameters to be applied to the processing step in the substrate processing device 100.

[0010] When manufacturing substrates such as semiconductor wafers, deformations such as warping and bending may occur in the substrate. 3D stacking technology is essential for improving semiconductor integration density, but it is known that as stacking increases, not only concentric warping (e.g., bowl-shaped warping) but also warping in non-concentric shapes (e.g., saddle-shaped warping) can occur. Saddle-shaped warping makes the bonding process more difficult and leads to a deterioration in bonding accuracy.

[0011] The substrate processing apparatus 100 is a substrate processing apparatus for correcting arbitrary deformations, including such saddle-shaped warping. The substrate processing apparatus 100 corrects the deformation of the substrate to be processed by performing the following steps: (1) a step of inverting the front and back sides of the substrate to be processed; (2) a step of forming a photocrosslinked stress film on the back surface of the substrate to be processed; and (3) a step of exposing the formed stress film and applying heat treatment as necessary.

[0012] Figure 1 shows an example of a substrate to be processed, in which semiconductor devices such as transistors and memory cells are formed on the surface of a base substrate, and deformation has occurred. The substrate processing apparatus 100 first inverts the substrate to be processed in order to deposit a stress film on the back surface of the substrate. Next, the substrate processing apparatus 100 deposits a photocrosslinked stress film on the back surface of the inverted substrate. For example, the substrate processing apparatus 100 deposits a photocrosslinked stress film by depositing a SiN film on the back surface of the substrate using chemical vapor deposition (CVD). Alternatively, inorganic materials such as SiO2 films, SiOC films, and metal films containing oxide nitrides can be used. Instead of inorganic materials, organic materials such as polyimide resins and epoxy resins may be used. The thickness of the stress film is adjusted by the deposition conditions such as the coater stage temperature, gas atmosphere, and rotation speed.

[0013] Next, the substrate processing apparatus 100 exposes the stress film deposited on the substrate. The substrate processing apparatus 100 partially crosslinks the stress film by using an appropriate mask pattern during exposure, thereby changing the thermal shrinkage rate. This allows for localized changes in the in-plane stress distribution during the subsequent baking process, and can correct deformations that occur in the substrate. Exposure conditions include the mask pattern described above, as well as the dose, exposure time, and resolution of the mask pattern.

[0014] Next, the substrate processing apparatus 100 heats the entire substrate in a heat treatment process (baking) and corrects the deformation due to the difference in thermal shrinkage rates between the substrate to be processed and the stress film. The mask pattern generates localized stress in the stress film, which can correct the deformation of the substrate. The baking conditions include the bake temperature, stage temperature, ambient temperature, moisture content, and bake time. Note that baking is not mandatory, and the deformation of the substrate may be corrected by forming the stress film and exposure.

[0015] To achieve the desired correction, it is necessary to appropriately set various processing parameters, including the mask pattern. Traditionally, experiments were conducted according to the amount of deformation (warpage) to empirically determine which processing parameters were appropriate. This had the problem of requiring a large amount of experimental effort.

[0016] Therefore, the information processing device 200 according to Embodiment 1 predicts the amount of deformation of the substrate according to the given processing parameters, evaluates the prediction result of the amount of deformation, searches for processing parameters (experimental conditions) to be applied to the processing step for correcting the deformation of the substrate, and outputs information on candidate processing parameters obtained as a result of the search.

[0017] Figure 2 is a block diagram showing the internal configuration of the information processing device 200. The information processing device 200 is a dedicated or general-purpose computer and includes a control unit 201, a storage unit 202, a communication unit 203, an operation unit 204, a display unit 205, and the like.

[0018] The control unit 201 includes a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and the like. The ROM in the control unit 201 stores control programs that control the operation of each hardware component of the information processing device 200. The CPU in the control unit 201 reads and executes the control programs stored in the ROM and the computer programs described later stored in the memory unit 202, and controls the operation of the hardware components, thereby making the entire device function as the information processing device 200 of this disclosure. The RAM in the control unit 201 temporarily stores data used during the execution of calculations.

[0019] In this embodiment, the control unit 201 is configured to include a CPU, ROM, and RAM, but the configuration of the control unit 201 is not limited to the above. The control unit 201 may be one or more control circuits or processing circuits that include, for example, a GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), quantum processor, volatile or non-volatile memory, etc. Furthermore, the control unit 201 may include functions such as a clock that outputs date and time information, a timer that measures the elapsed time from the time a measurement start instruction is given to the time a measurement end instruction is given, and a counter that counts numbers.

[0020] The storage unit 202 is equipped with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The storage unit 202 stores various computer programs executed by the control unit 201 and various data used by the control unit 201.

[0021] The computer program (program product) stored in the memory unit 202 includes a conditional search program PG that causes the computer to perform the following processes: predict the amount of deformation of the substrate according to the given processing parameters, evaluate the prediction result of the amount of deformation, search for processing parameters to be applied to the processing steps for correcting the deformation of the substrate, and output information on candidate processing parameters obtained as a result of the search.

[0022] The conditional search program PG may be a single computer program or a group of programs composed of multiple computer programs. The conditional search program PG may be executed collaboratively by multiple computers. The conditional search program PG may partially utilize existing libraries.

[0023] A computer program including a conditional search program PG is provided, for example, on a non-temporary recording medium RM on which the computer program is recorded in a readable format. The recording medium RM is a portable memory such as a CD-ROM, USB memory, SD (Secure Digital) card, microSD card, or CompactFlash®. The control unit 201 reads various computer programs from the recording medium RM using a reading device (not shown in the figure) and stores the read computer programs in the storage unit 202. The computer programs stored in the storage unit 202 may also be provided by communication. In this case, the control unit 201 acquires the computer programs by communication via the communication unit 203 and stores the acquired computer programs in the storage unit 202.

[0024] The communication unit 203 is equipped with a communication interface for sending and receiving various types of data with an external device. The communication interface of the communication unit 203 can be a wired or wireless communication interface compliant with a communication standard such as LAN (Local Area Network). The external device may be a board processing unit 100 or a user terminal (not shown). When data to be transmitted is input from the control unit 201, the communication unit 203 transmits the data to the destination external device, and when data transmitted from the external device is received, the communication unit 203 outputs the received data to the control unit 201.

[0025] The operation unit 204 is equipped with operating devices such as a touch panel, keyboard, and switches, and accepts various inputs and operations from the user. The control unit 201 acquires information input through the operation unit 204 and performs appropriate control based on the various operation information provided by the operation unit 204.

[0026] The display unit 205 is equipped with a display device such as a liquid crystal monitor or an organic EL (Electro-Luminescence) monitor, and displays information that should be notified to the user or others in response to instructions from the control unit 201.

[0027] The information processing device 200 may be a single computer, or it may be a computer system composed of multiple computers and peripheral devices. Furthermore, the information processing device 200 may be a virtual machine with a virtualized physical form, or it may be a cloud. In this embodiment, the information processing device 200 is a separate and independent device from the substrate processing device 100, but it may also be a device located inside the substrate processing device 100.

[0028] Figure 3 is an explanatory diagram illustrating the overview of the processing performed by the information processing device 200. The information processing device 200 includes a prediction model PM that predicts the amount of deformation of the substrate from given processing parameters (input parameters), and an optimization algorithm OA that evaluates the deformation prediction results and searches for processing parameters that should be applied to the processing step.

[0029] The prediction model PM uses a surrogate model that has been pre-trained on the relationship between the processing parameters used in the processing steps to correct substrate deformation and the amount of deformation after the processing steps. When the input parameter to the model is a mask pattern, both the mask pattern and the amount of deformation before and after the processing steps are represented as numerical data inside a two-dimensional circle. These numerical data can be expressed as a superposition of Zernike polynomials. Therefore, the correspondence between the Zernike component representing the amount of deformation after processing (predicted value) and the Zernike component representing the amount of deformation and mask pattern before processing is expressed mathematically as a linear relationship Y = AX + B. Here, Y is the Zernike component representing the amount of deformation after processing, X is the Zernike component representing the amount of deformation and mask pattern before processing, and A and B are coefficients.

[0030] Coefficients A and B are determined using data obtained by conducting preliminary experiments with the substrate processing apparatus 100. In the preliminary experiments, multiple mask patterns are set, and the amount of deformation of the substrate before processing and the amount of deformation of the substrate after performing the processing step using each mask pattern are measured. Coefficients A and B are determined by fitting the above formula to the Zernike component representing the amount of deformation after processing (measured value) and the Zernike component representing the amount of deformation before processing (measured value) and the mask pattern. After coefficients A and B are determined, the amount of deformation after processing can be predicted by inputting the amount of deformation before processing and the set mask pattern into the surrogate model represented by the above formula.

[0031] When considering processing parameters other than the mask pattern, such as film thickness and bake time, a surrogate model can be created by formulating the equations by separating them into a Zernike component term that represents processing parameters that linearly deform the whole (film thickness, bake time, etc.) and a Zernike component term that represents processing parameters that affect only specific components (mask amount), and then fitting them using the same procedure.

[0032] Furthermore, if the amount of data available is small, it is necessary to create a surrogate model for each deformation pattern. In this case, linear programming methods such as D-optimization can be used to efficiently create surrogate models.

[0033] The information processing device 200 evaluates the amount of deformation of the substrate predicted by the prediction model PM using an evaluation function EF. The evaluation function EF uses a function that calculates the difference (error) between the predicted deformation amount and the desired deformation amount. The desired deformation amount can be arbitrarily set by the user. Generally, the desired deformation amount is set to zero at all points within a two-dimensional circle. Alternatively, it is also possible to set it to minimize only a specific component (for example, concentric components).

[0034] The information processing device 200 modifies the processing parameters using the optimization algorithm OA based on the output of the evaluation function EF. In this embodiment, a multi-objective optimization algorithm such as TPE (Tree-structured Parzen Estimator) or MOTPE (Multi-Objective TPE) is used as the optimization algorithm OA. By using the multi-objective optimization algorithm OA, for example, it is possible to reduce the predicted deformation amount while reducing specific components of the deformation amount (such as concentric bowl components and non-concentric saddle components). In multi-objective optimization, when the priority of indicators that have at least some trade-off relationships is changed, there are multiple optimal solutions, which are called Pareto optimal solutions (or Pareto solutions).

[0035] The information processing device 200 inputs the processing parameters modified by the optimization algorithm OA back into the prediction model PM. The information processing device 200 repeatedly performs prediction by the prediction model PM, evaluation of the prediction results using the evaluation function EF, and search for a Pareto solution using the optimization algorithm OA. The number of repetitions is set in advance. Alternatively, the above calculations may be repeated until the evaluation by the evaluation function EF falls below a threshold.

[0036] The information processing device 200 repeatedly performs prediction using the prediction model PM, evaluation using the evaluation function EF, and search for Pareto solutions using the optimization algorithm OA, and outputs information on candidate processing parameters obtained as a result of the search. The optimization algorithm OA provides the modified processing parameters, and the prediction model PM provides the predicted deformation amount when the modified processing parameters are used, so the information processing device 200 displays them on the display unit 205.

[0037] Figure 4 is a flowchart showing the procedure for searching for experimental conditions performed by the information processing device 200. The control unit 201 of the information processing device 200 reads and executes the condition search program PG stored in the storage unit 202, thereby performing the following processing.

[0038] The control unit 201 receives input of processing parameters from the user through the operation unit 204 (step S101). The processing parameters received in step S101 include the film deposition conditions in the film deposition process performed by the substrate processing apparatus 100 and the exposure conditions in the exposure process. The film deposition conditions include coater stage temperature, gas atmosphere, rotation speed, etc. The exposure conditions include mask pattern, dose amount, exposure time, resolution of the mask pattern, etc. The mask pattern is input as an image in which the exposed areas are drawn with white pixels and the unexposed areas with black pixels. The processing parameters received in step S101 may further include baking conditions in the heat treatment process (baking) performed by the substrate processing apparatus 100. The baking conditions include bake temperature, stage temperature, atmosphere temperature, moisture content, bake time, etc.

[0039] Furthermore, the control unit 201 may accept information regarding the substrate to be processed, information regarding the material used for the stress film, equipment information for the substrate processing apparatus 100, and equipment information for the substrate processing apparatus (film deposition apparatus, etching apparatus, exposure apparatus, polishing apparatus, annealing apparatus, cleaning apparatus, bonding apparatus, etc.) used before and after correction as processing parameters.

[0040] The control unit 201 inputs processing parameters to the prediction model PM and predicts the amount of deformation of the substrate after applying the processing parameters and correcting the deformation (step S102). The prediction model PM is assumed to have been created in advance according to the type of processing parameters used for prediction using the method described above. The predicted deformation amount is represented as numerical data inside a two-dimensional circle. The control unit 201 may predict the deformation amount of the entire substrate, or it may predict the deformation amount for each specific component. For example, the control unit 201 may estimate the concentric components of the deformation amount (hereinafter referred to as the bowl component), or it may estimate components other than the concentric circles (hereinafter referred to as the saddle component). The initial processing parameters given to the prediction model PM are the processing parameters received in step S101, but the processing parameters given from the second time onward are the processing parameters after they have been modified by the optimization algorithm OA.

[0041] The control unit 201 evaluates the prediction result obtained in step S102 using the evaluation function EF (step S103). If the evaluation function EF is set as a function that calculates the difference between the predicted deformation and the desired deformation, the control unit 201 evaluates the predicted deformation by calculating the difference between the predicted deformation and the desired deformation. The desired deformation is arbitrarily set by the user. Alternatively, the desired deformation may be set in advance.

[0042] The control unit 201 determines whether or not to terminate the search for experimental conditions (step S104). For example, the control unit 201 determines to terminate the search if the evaluation in step S103 falls below a threshold. Here, the control unit 201 may also determine to terminate the search if the overall average value of the predicted deformation amount falls below a threshold, or if a specific component of the predicted deformation amount falls below a threshold. The control unit 201 may also count the number of times the processing in steps S102 to S105 is repeated, and determine to terminate the search when the count reaches a set number.

[0043] If it is determined in step S104 not to terminate the search (S104: NO), the control unit 201 modifies the processing parameters based on the output of the evaluation function EF (step S105). After modifying the processing parameters, the control unit 201 returns to step S102 and repeats the process from S102 to S105. A multi-objective optimization algorithm (optimization algorithm OA) such as TPE or MOTPE is used to modify the processing parameters. The optimization algorithm OA allows for the search of processing parameters so that the predicted deformation amount is optimal while modifying the processing parameters, including the mask pattern. In this embodiment, there are multiple indicators that have a trade-off relationship with each other, such as the predicted deformation amount of the entire substrate, the bowl component of the predicted deformation amount, and the saddle component of the predicted deformation amount, so multiple optimal solutions (Pareto solutions) can be obtained depending on which indicator is prioritized.

[0044] When it is determined in step S104 that the search is to be terminated (S104: YES), the control unit 201 outputs information on candidates for processing parameters obtained as search results (step S106). Since the control unit 201 repeatedly executes the processes of steps S102 to S105 to search for a plurality of optimal solutions, a plurality of processing parameters are obtained in the process. The control unit 201 causes the display unit 205 to display the information on the obtained plurality of processing parameters as candidates for processing parameters to be applied to the processing steps for correcting the deformation of the substrate in the substrate processing apparatus 100. Alternatively, the control unit 201 may notify the user's terminal device of the information on the candidates for processing parameters through the communication unit 203.

[0045] FIG. 5 is a schematic diagram showing an output example of candidates for processing parameters. The output example in FIG. 5 shows an example in which candidates for processing parameters are displayed on the display unit 205 of the information processing apparatus 200. In the embodiment, since multi-objective optimization is adopted, a plurality of candidates for processing parameters are obtained. In the example of FIG. 5, as examples of candidates for processing parameters, a mask pattern (proposed mask) and the rotation speed of the coater stage are shown. The processing parameters displayed on the display unit 205 may be only the processing parameters corrected by the information processing apparatus 200 based on the evaluation result of the evaluation function EF, or may include the processing parameters initially input by the user himself / herself. In addition to the mask pattern and the rotation speed of the coater stage, the information processing apparatus 200 may display the temperature of the coater stage, the gas atmosphere, etc., which are film formation conditions. Further, the information processing apparatus 200 may display the dose amount, exposure time, resolution of the mask pattern, etc., which are exposure conditions. Furthermore, the information processing apparatus 200 may display the baking temperature, stage temperature, atmosphere temperature, moisture content, baking time, etc., which are baking conditions.

[0046] Since the information processing apparatus 200 corrects processing parameters and predicts the amount of substrate deformation when the corrected parameters are applied, the prediction results can be displayed together. FIG. 5 shows an example in which the prediction error (saddle component) and the bowl maximum value when the deformation of the substrate is corrected by applying each processing parameter are displayed together. The saddle component of the prediction error represents the saddle component among the values obtained by subtracting the desired deformation amount from the predicted deformation amount. The bowl maximum value represents the maximum value of the bowl component among the predicted deformation amounts. The information processing apparatus 200 may display the overall predicted deformation amount in addition to or instead of the saddle component of the prediction error, or may display the bowl component of the prediction error.

[0047] The user can refer to the prediction results of the amount of substrate deformation displayed on the display unit 205 and select the processing parameters to be applied to the processing steps when correcting the deformation of the substrate using the substrate processing apparatus 100.

[0048] After outputting information on candidates for processing parameters in step S106, the control unit 201 receives a selection of the processing parameters desired by the user (step S107). The control unit 201 may receive a selection of processing parameters on a display screen as shown in FIG. 5. Alternatively, the control unit 201 may select the processing parameters to be applied to the processing steps from among the candidate processing parameters. In this case, it is preferable that selection criteria such as whether to select the processing parameters so that the overall deformation amount is minimized or whether to select the processing parameters so that the saddle component of the prediction error is minimized are set in advance.

[0049] The control unit 201 outputs a control command to the substrate processing apparatus 100 so as to perform a processing step using the processing parameters selected in step S107 (step S108). By outputting a control command from the information processing apparatus 200 to the substrate processing apparatus 100, the substrate can be corrected within the range of the prediction error.

[0050] In this flowchart, steps S107 to S108 are not mandatory, and the process may end with step S106, which presents processing parameters to the user.

[0051] As described above, in Embodiment 1, the amount of deformation of the substrate is predicted according to the given processing parameters, the predicted deformation amount is evaluated, processing parameters (experimental conditions) to be applied to the processing step for correcting the deformation of the substrate are searched for, and information on candidate processing parameters obtained as a result of the search is output.

[0052] In Embodiment 1, a surrogate model was exemplified as the prediction model PE. Alternatively, the prediction model PE may be a machine learning model. Figure 6 is a schematic diagram showing a modified example of the prediction model PE. The prediction model PE shown in Figure 6 is a machine learning model that, when processing parameters are input, is learned to output the deformation amount after correction of the substrate deformation in the substrate processing device 100. Existing neural networks such as U-Net (U-Shaped Network) and CNN (Convolutional Neural Network) can be used as the learning model. Such a learning model is generated by collecting processing parameters, including the mask pattern used in a preliminary experiment using the substrate processing device 100, and measured values ​​of the corrected substrate deformation amount, and then learning with an existing learning algorithm using the collected dataset as training data.

[0053] Furthermore, the mask pattern used in the training data does not need to be a binary mask pattern represented by white and black pixels; it may be a mask pattern converted to continuous values. A mask pattern converted to continuous values ​​can be generated, for example, by summing Zernike polynomials. A learning model trained using such training data will yield a mask pattern in the form of continuous values. Therefore, the information processing device 200 can convert the continuous mask pattern obtained from the learning model to binary using a Gaussian filter or the like, and then present the converted mask pattern to the user.

[0054] The data input to the learning model may include not only the mask pattern but also the uncorrected substrate deformation. The mask pattern and the uncorrected substrate deformation are input to the learning model as image data. The uncorrected substrate deformation may be input separately as bowl components and saddle components.

[0055] Furthermore, the data input to the learning model may include deposition conditions such as coater stage temperature, gas atmosphere, and rotation speed; exposure conditions such as mask pattern, dose amount, exposure time, and mask pattern resolution; and baking conditions such as bake temperature, stage temperature, atmosphere temperature, moisture content, and bake time. In addition, the data input to the learning model may include data from the immediately preceding recipe (temperature time series data and gas type data used in the immediately preceding recipe), and may also include data from various devices including the substrate processing apparatus 100.

[0056] The information processing device 200 can present the user with multiple candidate processing parameters by repeatedly performing calculations to predict substrate deformation using a prediction model PM based on a learning model, and calculations to modify processing parameters according to the evaluation of the prediction results and search for experimental conditions using an optimization algorithm OA. When a machine learning learning model is used as the prediction model PE, a large amount of experimental data is required during training, but the expressive power increases, allowing for the proposal of more processing parameters and the reproduction of finer mask pattern shapes.

[0057] Figure 6 shows a learning model that, when given data such as a mask pattern, is trained to output the deformation amount after correction for substrate deformation in the substrate processing apparatus 100. Alternatively, a learning model may be used that, when given data on processing parameters including the mask pattern and data on the predicted deformation amount, outputs the processing parameters to be applied in the processing steps of the substrate processing performed by the substrate processing apparatus 100. In this case, since the learning model directly outputs the processing parameters, calculations by the optimization algorithm OA are unnecessary.

[0058] (Embodiment 2) Embodiment 2 describes a configuration that accepts constraints on processing parameters. The overall configuration of the experimental condition generation system and the internal configurations of the substrate processing device 100 and the information processing device 200 are the same as in Embodiment 1, so their description will be omitted.

[0059] Figure 7 is a schematic diagram showing an example of an input screen for accepting constraints on processing parameters. The control unit 201 of the information processing device 200 displays an input screen like the one shown in Figure 7 on the display unit 205 when accepting processing parameters. The input screen accepts constraints such as the generation of the device formed on the substrate, the material of the stress film, the type of bonding apparatus, the type of film deposition apparatus, the saddle shape of the corrected substrate, the maximum bowl value, the rotation speed of the coater stage, the temperature of the coater stage, and constraints on the mask pattern.

[0060] The input screen illustrated in Figure 7 is a user interface screen equipped with UI (User Interface) components such as pull-down menus, text boxes, image input areas, and operation buttons. The device generation, stress film material, bonding equipment type, and film deposition equipment type can be selected using pull-down menus. In the example in Figure 7, the bonding equipment type and film deposition equipment type can be selected, but it may also be possible to select types such as etching equipment, exposure equipment, polishing equipment, baking equipment, and cleaning equipment.

[0061] For the maximum bowl size, the rotation speed of the coater stage, the temperature of the coater stage, and constraints on the mask pattern, the user can specify the desired numerical value (or range of values) via a text box. Alternatively, the information processing device 200 may communicate with the substrate processing device 100 to obtain available numerical values ​​(or ranges of values) from the substrate processing device 100 and display them in the text box. By accepting constraints on the mask pattern, it becomes possible to adjust the amount of masking or to choose not to place (or to place) masks on edges or centers.

[0062] Regarding the saddle shape of the corrected substrate, the user can input the desired saddle shape of the processed substrate in the image input area. Since it is preferable that the processed substrate does not have a saddle shape, nothing is often drawn in the image input area.

[0063] When the user presses the optimization execution button, which is located as an operation button on the input screen, the control unit 201 of the information processing device 200 executes the optimization process. Specifically, it performs the deformation amount prediction, evaluation of the prediction results, and the search for experimental conditions using the optimization algorithm OA (processing steps S102 to S106 in the flowchart of Figure 4), as described in Embodiment 1. If constraint conditions for processing parameters are received on the input screen, the control unit 201 only needs to search for processing parameters within the range of the received constraint conditions.

[0064] As described above, in Embodiment 2, constraints on processing parameters can be accepted, so processing parameters can be proposed within the range desired by the user.

[0065] (Embodiment 3) Embodiment 3 describes a configuration that accepts user modifications to the presented processing parameters. The overall configuration of the experimental condition generation system and the internal configurations of the substrate processing device 100 and the information processing device 200 are the same as in Embodiment 1, so their description will be omitted.

[0066] Figure 8 is a schematic diagram showing an example of a modification screen that accepts modifications to processing parameters. Figure 8 shows an example where the information processing device 200 accepts user modifications to the original mask pattern presented. The mask pattern is presented as an image in which, for example, the exposed areas are drawn with white pixels and the unexposed areas with black pixels. The information processing device 200 modifies the mask pattern by accepting modifications to the image. The example in Figure 8 shows an example where a modification is accepted to reduce the inward bulge of the mask at the edges and to expand the exposed area.

[0067] When the control unit 201 of the information processing device 200 receives a modification to the processing parameters, including the mask pattern, it recalculates the substrate deformation amount according to the modified processing parameters, evaluates the deformation amount, and re-searches for the processing parameters. Specifically, the control unit 201 re-executes the processes shown in steps S102 to S106 of the flowchart in Figure 4. The example in Figure 8 shows that the prediction error and the maximum bowl value have been modified in accordance with the modification of the mask pattern.

[0068] Figure 8 shows an example where the prediction error and bowl maximum value were corrected by modifying the mask pattern. However, the system may also accept modifications to other processing parameters (e.g., the rotation speed of the coater stage) and reflect these changes in the mask pattern and prediction error.

[0069] As described above, in Embodiment 3, user modifications to the presented processing parameters can be accepted, and the accepted modifications can be reflected in the prediction error, etc.

[0070] (Embodiment 4) Embodiment 4 describes a configuration for displaying a mask pattern similar to the original mask pattern. The overall configuration of the experimental condition generation system and the internal configurations of the substrate processing device 100 and the information processing device 200 are the same as in Embodiment 1, so their description will be omitted.

[0071] The information processing device 200 according to Embodiment 4 acquires the applied processing parameters and data on the amount of deformation after processing (measured data) from the substrate processing device 100 after the substrate processing device 100 has performed a processing step to correct the deformation of the substrate, and stores them in the storage unit 202 as a database. The amount of deformation after processing can be measured using existing measuring devices such as a wafer optical inspection system.

[0072] When the information processing device 200 receives new processing parameters (mask patterns) from a user for the purpose of exploring experimental conditions, it may search the database for processing parameters similar to the received processing parameters and present them to the user. A mask pattern set as one of the processing parameters is, for example, drawn as an image in which the exposed areas are drawn as white pixels and the unexposed areas as black pixels. Therefore, it is possible to read mask patterns similar to the mask pattern set by the user from the database using existing methods for determining the similarity between images. As an existing method for determining the similarity between images, for example, a method can be used to evaluate similarity by extracting feature vectors from the target image and comparing the distance and direction of the extracted feature vectors. Alternatively, a method can be used to evaluate similarity by comparing the coefficients of the Zernike components described above.

[0073] Figure 9 is a schematic diagram showing an example of displaying processing parameters extracted from the database. In the example in Figure 9, the original mask pattern and rotation speed are processing parameters set by the user for the purpose of exploring experimental conditions. The information processing device 200 predicts the amount of deformation of the substrate using the processing parameters set by the user. In the example in Figure 9, the saddle component and the maximum bowl value of the prediction error are displayed as prediction results.

[0074] Furthermore, the information processing device 200 searches the database for mask patterns similar to the mask pattern set by the user. The information processing device 200 displays the searched mask patterns and various data stored in the database associated with those mask patterns as search results. In the example in Figure 9, the two searched mask patterns are displayed along with the deformation measurement result (saddle component), the bowl maximum value, and the rotation speed of the coater stage.

[0075] In the example in Figure 9, two search results are displayed, but the number of search results is not limited to two. The configuration may display only the single most similar result, or it may display three or more search results.

[0076] The user can refer to past experimental results (such as processing parameters retrieved from a database) presented by the information processing device 200 and select new processing parameters to be applied. The information processing device 200 may also output control commands to the substrate processing device 100 to perform a processing step using the processing parameters selected by the user, similar to the first embodiment.

[0077] As described above, in Embodiment 4, past experimental results similar to the processing parameters set by the user can be searched from the database and presented to the user.

[0078] (Embodiment 5) In the embodiment, a mask pattern is prepared to correct deformation that occurs in the substrate, and a crosslinked region and an uncrosslinked region are formed by exposure, so a step may occur between the regions. Due to the effect of this step, there is a concern that leakage may occur when the substrate is held in the vacuum chuck during the post-correction process, making it impossible to carry out the process. Embodiment 5 proposes a mask pattern that suppresses leakage and allows vacuum adsorption to be carried out without problems even when there is a step.

[0079] Figure 10 is a schematic diagram showing a first example of a mask pattern in Embodiment 5. The mask pattern illustrated in Figure 10 is a mask pattern that, in addition to a mask pattern for correcting deformation, sets the outermost area as an area that is not exposed. By using such a mask pattern, steps at the outermost edge can be eliminated. This improves the adhesion of the edge portion, which is important for vacuum adsorption, and reduces leakage not only at the edge portion but also overall.

[0080] Figure 11 is a schematic diagram showing a second example of the mask pattern in Embodiment 5. The mask pattern illustrated in Figure 11 is a mask pattern set to expose the entire outermost area, in addition to a mask pattern for correcting deformation. Even when using such a mask pattern, steps at the outermost edge can be eliminated. Similar to the example in Figure 10, the adhesion of the edge portion, which is important for vacuum adsorption, can be improved, and leakage can be reduced not only at the edge portion but also overall.

[0081] Figure 12 is a schematic diagram showing a third example of a mask pattern in Embodiment 5. The mask pattern illustrated in Figure 12 is a mask pattern in which the areas to be exposed are formed as a pixel pattern. Here, the pixel pattern is a pattern in which a first area to be exposed and a second area not to be exposed are arranged alternately and continuously. In the example of Figure 12, the first area and the second area are rectangular regions of the same area, forming a checkerboard pattern overall, but the first area and the second area do not have to have the same area, and do not have to be rectangular regions. When a pixel pattern is formed, the slopes of the exposed areas overlap, which reduces the actual step difference and can reduce leakage.

[0082] Figure 13 is a schematic diagram showing a fourth example of the mask pattern in Embodiment 5. The mask pattern illustrated in Figure 13 is a mask pattern in which a pixel pattern is arranged around a mask pattern for correcting deformation. By using such a mask pattern, as in the example in Figure 12, the slopes of the exposed areas overlap, thereby reducing the actual step difference and reducing leakage.

[0083] Figure 14 is a schematic diagram showing a fifth example of the mask pattern in Embodiment 5. The mask pattern illustrated in Figure 14 is a mask pattern in which a mask pattern for correcting deformation is formed as a pixel pattern. By using such a mask pattern, as in the example in Figure 12, the slopes of the exposed areas overlap, thereby reducing the actual step difference and reducing leakage.

[0084] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended.

[0085] For example, the processing steps for correcting substrate deformation are not limited to those disclosed in the embodiments and can be arbitrarily designed. The processing steps may include a film deposition step for forming a stress film on the substrate, but may not include an exposure step for exposing the stress film formed on the substrate, or a heating step for heating the stress film. In this case, the information processing device 200 can use the method disclosed in the embodiments to search for film deposition conditions in the film deposition step as processing parameters and output information on candidate processing parameters obtained as a result of the search. Furthermore, the processing steps may include a film deposition step and an exposure step, but may not include a heating step. In this case, the information processing device 200 can use the method disclosed in the embodiments to search for film deposition conditions in the film deposition step and exposure conditions in the exposure step as processing parameters, and output information on candidate processing parameters obtained as a result of the search. Moreover, the processing steps may include a film deposition step and a heating step, but may not include an exposure step. In this case, the information processing device 200 can use the method disclosed in the embodiment to search for the film deposition conditions in the film deposition process and the heating conditions in the heating process as processing parameters, and output information on candidate processing parameters obtained as a result of the search.

[0086] Furthermore, the matters described in each embodiment can be combined with each other. The independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. The claims may be described in a form in which claims refer to two or more other claims (multi-claim form), or in a form in which multi-claims refer to at least one multi-claim (multi-multi-claim form).

[0087] 100 Substrate processing device 200 Information processing device 201 Control unit 202 Storage unit 203 Communication unit 204 Operation unit 205 Display unit PG Condition search program RM Recording medium

Claims

1. A computer program that outputs information on processing parameters to be applied to a processing step for correcting deformation of a substrate, wherein the computer receives constraints in the processing step, predicts the amount of deformation of the substrate according to the given processing parameters, evaluates the prediction result of the amount of deformation, searches for processing parameters to be applied to the processing step under the accepted constraints, and outputs information on candidate processing parameters obtained as a result of the search.

2. The computer program according to claim 1, wherein the processing step includes a film deposition step of forming a stress film on the substrate, and the processing parameters include film deposition conditions in the film deposition step.

3. The computer program according to claim 2, wherein the processing step further includes an exposure step of exposing a stress film formed on the substrate, and the processing parameters further include exposure conditions in the exposure step.

4. The computer program according to claim 3, which causes the computer to perform a process of displaying side by side a plurality of mask patterns obtained as a result of the search and the amount of deformation of the substrate when the processing step is performed using each mask pattern.

5. The computer program according to claim 4, which causes the computer to perform a process that displays the processing parameters used in the processing step.

6. The computer program according to claim 4, which causes the computer to perform a process of accepting a modification to the displayed mask pattern, predicting the amount of deformation of the substrate according to the modified mask pattern, evaluating the predicted amount of deformation, and re-searching for processing parameters to be applied to the processing step.

7. The computer program according to claim 5, which causes the computer to perform a process of accepting a modification to the displayed processing parameters, predicting the amount of deformation of the substrate according to the modified processing parameters, evaluating the predicted amount of deformation, and re-searching for processing parameters to be applied to the processing step.

8. The computer program according to claim 1, which causes the computer to perform a process that displays the result of a processing step when processing parameters similar to the searched processing parameters are applied, by referring to a storage unit that stores processing parameters in association with the results of a processing step performed by applying the processing parameters.

9. The computer program according to claim 1, which causes the computer to perform a process of receiving a selection of processing parameters to be applied to the processing step from the outputted candidates of processing parameters, and outputting a control command to the device to perform the processing step by applying the selected processing parameters.

10. An information processing method for outputting information on processing parameters to be applied to a processing step for correcting deformation of a substrate, the method comprising: receiving constraints in the processing step; predicting the amount of deformation of the substrate according to the given processing parameters; evaluating the result of the deformation prediction; searching for processing parameters to be applied to the processing step under the accepted constraints; and outputting information on candidate processing parameters obtained as a result of the search, all performed by a computer.

11. An information processing device that outputs information on processing parameters to be applied to a processing step for correcting deformation of a substrate, comprising at least one processor, wherein the processor receives constraints in the processing step, predicts the amount of deformation of the substrate according to the given processing parameters, evaluates the prediction result of the amount of deformation, searches for processing parameters to be applied to the processing step under the accepted constraints, and outputs information on candidate processing parameters obtained as a result of the search.

Images