Mask key dimension data analysis processing method and device
By analyzing key dimension data of the mask feature pattern, determining the effective grayscale threshold, and training the optical proximity effect model, the problem of incomplete data utilization in the OPC model was solved, the inclusion accuracy of photoresist morphology was improved, and higher fitting accuracy was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI INTEGRATED CIRCUIT EQUIPMENT & MATERIALS INDUSTRY INNOVATION CENTER CO LTD
- Filing Date
- 2022-12-30
- Publication Date
- 2026-06-02
AI Technical Summary
The existing OPC model does not fully utilize data, has low fitting accuracy, and lacks vertical topographic information, resulting in low data utilization when the complexity and accuracy requirements of the photolithography process increase.
By collecting key dimension data of the feature pattern of the photomask, measuring it with a measurement device, calculating the difference under different gray level thresholds, determining the effective gray level threshold, and training the optical proximity effect model with the effective key dimension data, the accurate optical proximity effect model is obtained, and finally the morphology of the photoresist is obtained.
It improves the fitting accuracy of the OPC model, solves the problem of incomplete data utilization, and enhances the accuracy of incorporating photoresist morphology into the model.
Smart Images
Figure CN116385346B_ABST
Abstract
Description
Technical Field
[0001] This application relates to semiconductor technology, and more particularly to a method and apparatus for analyzing and processing key dimension data of a photomask. Background Technology
[0002] Photolithography is one of the most critical processes in chip manufacturing, transferring designed patterns onto silicon wafers using a photolithography imaging system. As chip sizes continue to shrink, the exposed patterns on the silicon wafer become distorted. Therefore, before photolithography manufacturing of chips at 90nm or even below 180nm, an algorithm model called Optical Proximity Conformity (OPC) must be used to optimize the exposed images. The accuracy of the OPC model therefore plays a crucial role in the photolithography process.
[0003] In current conventional OPC modeling processes, each measurement image is quantified into a single numerical value for model building. A large number of measurement images are not fully utilized, and only the dimensional information of the feature patterns within the silicon wafer plane is obtained, without vertical morphological information. Furthermore, with the increasing complexity of photolithography processes and the increasing requirements for fitting accuracy, more detailed data analysis and processing of the measurement images are needed to improve data utilization and incorporate photoresist morphology into the model. Summary of the Invention
[0004] This application provides a method and apparatus for analyzing and processing key dimension data of a mask, in order to solve the problems of incomplete data utilization and low fitting accuracy in existing OPC models.
[0005] On the one hand, this application provides a method for analyzing and processing key dimension data of a mask, including:
[0006] Acquire mask feature images, which include the mask feature images to be trained and the mask images to be processed;
[0007] The mask feature pattern is measured using a measurement device to obtain the key dimension data of the mask feature pattern under different grayscale thresholds. After calculating the difference between the key dimension data of the same feature pattern, the difference is analyzed to determine the effective grayscale threshold of the same feature pattern.
[0008] The optical proximity effect model is trained using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model. The effective key size data is the key size data corresponding to the effective gray level threshold.
[0009] The effective key dimension data of the feature pattern of the mask to be processed is sent to the precise optical proximity effect model to obtain the key dimensions and morphology of the photoresist corresponding to the pattern of the mask to be processed.
[0010] On the other hand, this application provides a mask key dimension data analysis and processing apparatus, including:
[0011] The acquisition module is used to acquire mask feature images, which include mask feature images to be trained and mask feature images to be processed.
[0012] The measurement module is used to measure the feature pattern of the mask using a measurement device, obtain the key size data of the feature pattern of the mask under different gray level thresholds, calculate the difference between the key size data of the same feature pattern, analyze the difference, and determine the effective gray level threshold of the same feature pattern.
[0013] The training module is used to train the optical proximity effect model using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model. The effective key size data is the key size data corresponding to the effective gray level threshold.
[0014] The learning module is used to send the effective key size data of the feature pattern of the mask to be processed to the precise optical proximity effect model to obtain the photoresist morphology corresponding to the feature pattern of the mask to be processed.
[0015] In another aspect, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;
[0016] The memory stores computer-executed instructions;
[0017] The processor executes computer execution instructions stored in the memory to implement the method described above.
[0018] In another aspect, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the mask key dimension data analysis and processing method described above.
[0019] The mask key dimension data analysis and processing method and apparatus provided in this application measures the acquired mask feature image using a measurement device to obtain the key dimension data of the mask feature image under different gray level thresholds. After calculating the difference between the key dimension data of the same feature image, the difference is analyzed to determine the effective gray level threshold of the same feature image. Then, the effective key dimension data of the mask feature image to be trained and the acquired photoresist morphology to be trained are used to train an optical proximity effect model to obtain an accurate optical proximity effect model. Finally, the effective key dimension data of the mask feature image to be processed is sent to the model to obtain the photoresist morphology corresponding to the mask feature image to be processed. This can effectively improve the fitting accuracy of the OPC model and solve the problem of long weight setting time in the OPC model. Attached Figure Description
[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0021] Figure 1 This is a schematic diagram of a mask key dimension data analysis and processing architecture based on the embodiments of this application;
[0022] Figure 2 A flowchart illustrating the mask key dimension data analysis and processing method provided in the embodiments of this application;
[0023] Figure 3 A schematic diagram of the signaling interaction of the mask key dimension data analysis and processing method provided in the embodiments of this application;
[0024] Figure 4 A structural block diagram of the mask key dimension data analysis and processing device provided in the embodiments of this application;
[0025] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0026] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0027] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0028] Photolithography is one of the most critical processes in chip manufacturing, transferring designed patterns onto silicon wafers using a photolithography imaging system. As chip sizes continue to shrink, distortion occurs in the exposed patterns on the silicon wafer. Therefore, before photolithography manufacturing of chips at 90nm and even below 180nm, an algorithm model called Optical Proximity Convergence (OPC) must be used to optimize the exposed images. The accuracy of the OPC model is therefore crucial in the photolithography process. Currently, conventional OPC modeling processes quantize each measurement image into a single numerical value for model building, resulting in the underutilization of numerous measurement images. Furthermore, only the dimensional information of the feature pattern within the silicon wafer plane is obtained, without vertical morphological information. With the increasing complexity of photolithography processes and the increasing demands for fitting accuracy, more detailed data analysis and processing of the measurement images are needed to improve data utilization and incorporate the photoresist morphology into the model.
[0029] Figure 1 This is a schematic diagram of a mask key dimension data analysis and processing architecture based on the embodiments of this application. See also: Figure 1 As shown, it mainly includes: a scanning electron microscope (CD-SEM) measurement device 101 for feature size measurement, a data preprocessing server 102, and a model server 103. The CD-SEM measurement device 101 can transmit key dimension data of the mask feature pattern to the data preprocessing server 102, and the data preprocessing server 102 then sends the processed key dimension data of the mask feature pattern to the model server 103.
[0030] CD-SEM Measurement Equipment 101 is a scanning electron microscope that determines the boundary of a pattern based on the grayscale of an image and then calculates the linewidth.
[0031] Data preprocessing server 102 is a server specifically designed for data preprocessing. Data preprocessing refers to the necessary processing, such as reviewing, filtering, and sorting, performed on the collected data before it is classified or grouped.
[0032] The model server 103 stores various algorithm models. These algorithm models can be either mathematical models or simulation models. In this application, they specifically refer to relevant optical proximity effect models, precise optical proximity effect models, and other algorithm models.
[0033] Current methods for model building to analyze photomask feature patterns quantize only a single numerical value from each measurement image for model building, resulting in underutilization of a large number of CD-SEM images. Furthermore, the use of a single threshold only provides dimensional information of the feature pattern within the XY plane of the silicon wafer, lacking information in the Z direction. These problems have existed but were not prominent; however, with the continuous miniaturization of devices, the requirements for model accuracy are increasing. Therefore, this application proposes a photomask critical dimension data analysis and processing method that considers the three-dimensional effects of photoresist and fully utilizes measurement data.
[0034] Figure 2 A flowchart illustrating the mask key dimension data analysis and processing method provided in this application embodiment is shown below. Figure 2 As shown in the embodiments of this application, the mask key dimension data analysis and processing method includes:
[0035] S201. Acquire the feature image of the mask.
[0036] The mask feature images include not only the mask feature images to be trained, but also the mask feature images to be processed. The mask feature images to be trained refer to the mask feature images used for model training, while the mask feature images to be processed refer to the mask feature images used for the trained model, where the photoresist morphology corresponding to the original mask feature images has not yet been obtained.
[0037] The types, line widths, and periodic dimensions of the mask feature graphics cover the range of graphics at the current level in the actual product.
[0038] Preferably, the mask feature graphics include both one-dimensional and two-dimensional graphics.
[0039] Preferably, the mask feature graphics include dense graphics, semi-dense graphics, and isolated graphics.
[0040] This method receives different types of mask feature patterns, making the mask feature patterns available for analysis more comprehensive and improving the utilization rate of mask feature patterns.
[0041] S202. Measure the feature pattern of the mask using a measuring device to obtain the key dimension data of the feature pattern under different grayscale thresholds. Calculate the difference between the key dimension data of the same feature pattern, analyze the difference, and determine the effective grayscale threshold of the same feature pattern.
[0042] Each grayscale threshold has a corresponding key size data.
[0043] At least two grayscale thresholds should be set for the feature images of the training mask. The grayscale thresholds include one baseline threshold and multiple ordinary thresholds. There can be one or three ordinary thresholds, and the specific number of ordinary thresholds is not limited here.
[0044] Different thresholds will result in different key dimensions. The differences between these key dimensions are analyzed to determine whether the difference is greater than a preset difference. If it is greater, the effective grayscale threshold for the same feature graphic is the grayscale threshold of all grayscale values for the same feature graphic; if it is less than or equal to, the effective grayscale threshold for the same feature graphic is the baseline threshold for the same feature graphic.
[0045] Preferably, when setting different grayscale thresholds, they can be specifically set as outer low threshold, outer middle threshold, high threshold, inner middle threshold, and inner low threshold.
[0046] In one implementation, when measuring the linewidth of the photoresist, the outer middle threshold is used as the reference threshold.
[0047] In another implementation, when measuring the line width of a trench or hole, the inner middle threshold is used as the reference threshold.
[0048] This method expands and filters the grayscale thresholds, thereby increasing the utilization rate of key dimension data while ensuring its availability.
[0049] S203. The optical proximity effect model is trained using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model.
[0050] Among them, the effective key dimension data is the key dimension data corresponding to the effective grayscale threshold.
[0051] Before training the optical proximity effect model, it is necessary to set the fitting weights for key size data under different grayscale thresholds.
[0052] Among them, the fitting weight of key dimension data under the benchmark threshold is greater than or equal to the fitting weight of key dimension data under the ordinary threshold.
[0053] Before training the optical proximity effect model, the initial values of the locations of the photoresist morphology to be trained need to be set to be variable. The number of locations of the photoresist morphology to be trained is equal to the number of effective grayscale thresholds.
[0054] After obtaining the accurate optical proximity effect model, we can also obtain the functional correspondence between the location points of the photoresist morphology to be trained and the effective grayscale thresholds corresponding to the photoresist morphology to be trained.
[0055] This method optimizes the input data of the optical proximity effect model, making the learned results of the trained optical proximity effect model more accurate.
[0056] S204. Send the mask pattern data to be processed to the precise optical proximity model to obtain the key dimensions and morphology of the photoresist corresponding to the mask pattern.
[0057] The data preprocessing server 102 sends the mask data to be processed to the accurate optical proximity model on the model server 103, thereby obtaining the key dimensions and morphology of the photoresist corresponding to the pattern of the mask to be processed.
[0058] This method learns from the effective key size data to be processed using an accurate optical proximity effect model, making the learning results more accurate and effectively improving the fitting accuracy.
[0059] Figure 3 A schematic diagram of the signaling interaction for the mask key dimension data analysis and processing method provided in the embodiments of this application is shown below. Figure 3 As shown, combined with Figure 2 The mask key dimension data analysis and processing method provided in this application embodiment includes the following steps:
[0060] S301, CD-SEM measurement equipment 101 receives the mask feature image uploaded by the staff.
[0061] S302 and CD-SEM measurement equipment 101 measure the mask pattern to obtain key dimension data of the mask feature pattern under different grayscale thresholds.
[0062] S303, CD-SEM measurement device 101 sends key dimension data to data preprocessing server 102.
[0063] S304, Data preprocessing server 102 calculates the difference between key dimension data of the same feature graphic.
[0064] S305, Data preprocessing server 102 analyzes the difference to obtain the effective grayscale threshold of the same feature image.
[0065] S306. The data preprocessing server 102 sends the effective key size corresponding to the effective grayscale threshold of the feature image of the mask to be trained to the model server 103.
[0066] S307 and model server 103 train the optical proximity effect model to obtain an accurate optical proximity effect model.
[0067] S308, Data preprocessing server 102 sends the effective key dimensions corresponding to the effective grayscale threshold of the feature image of the mask to be processed to model server 103.
[0068] S309 and model server 103 use a precise optical proximity effect model to learn and obtain the morphology of the photoresist.
[0069] Figure 4 This is a structural block diagram of the mask key dimension data analysis and processing device provided in the embodiments of this application. For ease of explanation, only the parts relevant to the embodiments of this application are shown. See also Figure 4 As shown, the mask key dimension data analysis and processing device provided in this application embodiment includes: acquisition module 401, measurement module 402, training module 403 and learning module 404.
[0070] The acquisition module 401 is used to acquire feature images of the mask.
[0071] The measurement module 402 is used to measure the feature pattern of the mask using a measurement device, obtain the key dimension data of the feature pattern of the mask under different gray level thresholds, calculate the difference between the key dimension data of the same feature pattern, analyze the difference, and determine the effective gray level threshold of the same feature pattern.
[0072] Training module 403 is used to train the optical proximity effect model using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model.
[0073] The learning module 404 is used to send the effective key size data of the feature pattern of the mask to be processed to the accurate optical proximity effect model to obtain the photoresist morphology corresponding to the feature pattern of the mask to be processed.
[0074] The mask key dimension data analysis and processing device provided in this application measures the acquired mask feature image using a measurement device to obtain key dimension data of the mask feature image under different gray level thresholds. After calculating the difference between key dimension data of the same feature image, the effective gray level threshold of the same feature image is determined. Then, the effective key dimension data of the mask feature image to be trained and the acquired photoresist morphology to be trained are used to train the optical proximity effect model to obtain an accurate optical proximity effect model. Finally, the effective key dimension data of the mask feature image to be processed is sent to the model to obtain the photoresist morphology corresponding to the mask feature image to be processed. This can effectively improve the fitting accuracy of the OPC model and solve the problem of incomplete data utilization in the OPC model.
[0075] Figure 5See the schematic diagram of the electronic device provided in the embodiments of this application. Figure 5 As shown, the electronic device includes: a memory 501, a processor 502, and a computer program; wherein the computer program is stored in the memory 501 and configured to be executed by the processor 502. Figure 2 , Figure 3 The various steps. Processor 502 is used to implement... Figure 4 Each module.
[0076] The memory 501 and the processor 502 are connected via a bus 503.
[0077] For relevant instructions, please refer to the corresponding text. Figures 2 to 4 The relevant descriptions and effects of the steps in the corresponding embodiments are understood, and will not be elaborated on here.
[0078] This application also provides a computer-readable storage medium including computer code that, when run on a computer, causes the computer to perform actions such as... Figures 2 to 5 The method provided by any of the corresponding implementation methods.
[0079] This application also provides a computer program product, including program code, which, when a computer runs the computer program product, executes as follows: Figures 2 to 5 The method provided by any of the corresponding implementation methods.
[0080] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.
[0081] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.
Claims
1. A method for analyzing and processing key dimension data of a photomask, characterized in that, include: Acquire mask feature images, which include mask feature images to be trained and mask feature images to be processed; The mask feature pattern is measured using a measurement device to obtain the key dimension data of the mask feature pattern under different grayscale thresholds, and the difference between the key dimension data of the same feature pattern is calculated to determine whether the difference is greater than a preset difference. If it is greater than the threshold value, the effective grayscale threshold value of the same feature graphic is the grayscale threshold value of all grayscale values of the same feature graphic. If the effective grayscale threshold of the same feature graphic is less than or equal to the threshold of the same feature graphic, then the effective grayscale threshold of the same feature graphic is the reference threshold of the same feature graphic. The optical proximity effect model is trained using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model. The effective key size data is the key size data corresponding to the effective gray level threshold. The image data of the mask to be processed is sent to the precise optical proximity model to obtain the key dimensions and morphology of the photoresist corresponding to the image of the mask to be processed.
2. The method of claim 1, wherein, Before measuring the feature image of the mask to be trained using a measurement device, the following steps are included: At least two grayscale thresholds are set for the feature image of the mask to be trained, the grayscale thresholds including a baseline threshold and at least one ordinary threshold.
3. The method according to claim 1 or 2, characterized in that, Before training the optical proximity effect model using the effective key size data of the feature image of the training mask and the morphology of the training photoresist, the method further includes: Set the fitting weights for key dimension data under different grayscale thresholds. The fitting weights for key dimension data under the baseline threshold are greater than or equal to the fitting weights for key dimension data under the ordinary threshold.
4. The method of claim 3, wherein, Before training the optical proximity effect model using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, the following steps are also included: The initial value of the vertical position point of the photoresist morphology to be trained is set to be variable.
5. The method of claim 4, wherein, The number of vertical position points of the photoresist morphology to be trained is equal to the number of effective grayscale thresholds.
6. The method of claim 1 or 2, wherein, After obtaining the accurate optical proximity effect model, it also includes: The functional correspondence between the vertical position point of the photoresist morphology to be trained and the effective grayscale threshold corresponding to the photoresist morphology to be trained is obtained.
7. A mask key dimension data analysis and processing device, comprising: The acquisition module is used to acquire mask feature images, which include mask feature images to be trained and mask feature images to be processed. The measurement module is used to measure the feature pattern of the mask using a measurement device, obtain the key size data of the feature pattern of the mask under different gray level thresholds, and calculate the difference between the key size data of the same feature pattern, and determine whether the difference is greater than a preset difference. If it is greater than the threshold value, the effective grayscale threshold value of the same feature graphic is the grayscale threshold value of all grayscale values of the same feature graphic. If the effective grayscale threshold of the same feature graphic is less than or equal to the threshold of the same feature graphic, then the effective grayscale threshold of the same feature graphic is the reference threshold of the same feature graphic. The training module is used to train the optical proximity effect model using the effective key size data of the feature image of the mask to be trained and the morphology of the photoresist to be trained, so as to obtain an accurate optical proximity effect model. The effective key size data is the key size data corresponding to the effective gray level threshold. The learning module is used to send the effective key size data of the feature pattern of the mask to be processed to the precise optical proximity effect model to obtain the photoresist morphology corresponding to the feature pattern of the mask to be processed.
8. An electronic device, comprising: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the mask key dimension data analysis and processing method as described in any one of claims 1-6.