Model and its establishing method and system, compensation method, device and storage medium

Abstract

A model, a method and system for establishing the same, a compensation method, a device and a storage medium. The method comprises: providing a measurement structure comprising a plurality of measurement patterns, the measurement pattern being an etched pattern; selecting a sampling position of the measurement structure, part of the measurement pattern of the sampling position representing characteristic information of the measurement structure; obtaining a plurality of sampling images of the sampling position; obtaining a pattern contour of the measurement pattern corresponding to the sampling position according to the plurality of sampling images; obtaining a plurality of measurement positions on the pattern contour; obtaining an etching deviation value of the measurement position according to the pattern contour; and establishing an etching deviation compensation model according to the etching deviation value of the measurement position. The present application improves the precision of the etching deviation compensation model.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing, and in particular to a model and its creation method and system, a compensation method, an apparatus, and a storage medium. Background Technology

[0002] Traditional rule-based etch bias compensation methods have faced challenges in terms of accuracy and efficiency, as etch bias tables cannot cover a wide range of complex design configurations. Variable Etch Bias (VEB) models offer a promising automated approach for more accurate etch bias compensation. However, they are no longer the preferred choice due to their poor accuracy, large data requirements, and high measurement time costs. Summary of the Invention

[0003] The problem solved by the embodiments of the present invention is to provide an etching deviation compensation model and its establishment method and system, compensation method, equipment and storage medium, thereby improving the accuracy of the etching deviation compensation model.

[0004] To address the aforementioned problems, this invention provides a method for establishing an etching deviation compensation model, comprising: providing a measurement structure, including multiple measurement patterns, wherein the measurement patterns are etched patterns; selecting sampling positions of the measurement structure, wherein a portion of the measurement pattern at the sampling position represents the feature information of the measurement structure; acquiring multiple sampling images at the sampling positions; obtaining the graphic contour of the measurement pattern corresponding to the sampling position based on the multiple sampling images; acquiring multiple measurement positions on the graphic contour; acquiring the etching deviation value of the measurement position based on the graphic contour; and establishing an etching deviation compensation model based on the etching deviation value of the measurement position.

[0005] Optionally, selecting sampling locations for the measurement structure includes: acquiring multiple initial sampling locations of the measurement structure, wherein the measurement patterns of the multiple initial sampling locations are different; acquiring graphic feature information of the measurement patterns of the initial sampling locations; and selecting a portion of the initial sampling locations as sampling locations, wherein the sum of the graphic feature information of the portion of the initial sampling locations covers the feature information of the measurement structure.

[0006] Optionally, obtaining graphic feature information of the measurement pattern of the initial sampling position includes: obtaining an initial sampling image of the initial sampling position; setting multiple initial measurement positions on the measurement pattern of the initial sampling image; obtaining graphic feature values ​​of the multiple initial measurement positions; obtaining the distribution of graphic feature information corresponding to each initial sampling image based on the multiple graphic feature values; selecting a portion of the initial sampling positions as sampling positions, and selecting the initial sampling positions corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measurement structure based on the feature information distribution of the initial sampling images as sampling positions.

[0007] Optionally, obtaining the graphic feature value of the initial measurement position includes: setting model feature parameters based on the feature information of the measurement structure; obtaining the simulated exposure pattern corresponding to the measurement pattern; and obtaining the value of the model feature parameter at the initial measurement position as the graphic feature value based on the simulated exposure pattern.

[0008] Optionally, in obtaining the simulated exposure pattern corresponding to the measured pattern, based on the initial measurement position, the simulated exposure pattern corresponding to the measured pattern is obtained in the initial sampled image.

[0009] Optionally, based on multiple graphic feature values, the distribution of graphic feature information corresponding to each initial sampled image is obtained, including: establishing a corresponding coordinate system space based on model feature parameters, with the coordinate axes of the coordinate system space corresponding one-to-one with the model feature parameters; setting multiple initial measurement positions in the coordinate system space with the values ​​of the model feature parameters at the initial measurement positions as the coordinates of the corresponding coordinate axes, and obtaining spatial points that correspond one-to-one with the initial measurement positions; dividing the coordinate system space of each initial sampled image into multiple class regions based on the spatial point distribution of each initial sampled image, and obtaining the distribution of spatial points in multiple class regions as the distribution of graphic feature information.

[0010] Optionally, the K-means clustering method can be used to divide the coordinate space of each initial sampled image into multiple class regions based on the spatial point distribution of each initial sampled image.

[0011] Optionally, based on the feature information distribution of the initial sampled images, the initial sampling positions corresponding to multiple initial sampled images whose feature information distribution covers the feature information of the measured structure are selected as sampling positions. This includes: obtaining the number of class regions covered by the spatial point distribution of each initial sampled image; and selecting the initial sampling positions corresponding to multiple initial sampled images in descending order of the number of class regions corresponding to the initial sampled images as sampling positions, wherein the distribution of the spatial points in the multiple initial sampled images covers all class regions.

[0012] Optionally, obtaining the graphic contour of the measurement pattern corresponding to the sampling position based on multiple sampled images includes: determining the corresponding position of the measurement pattern in the sampled images; obtaining the sampling contour of the measurement pattern at the corresponding position in multiple sampled images; and obtaining the graphic contour of the measurement pattern at the sampling position corresponding to the sampled images based on the sampling contours of multiple sampled images.

[0013] Optionally, determining the corresponding position of the measurement pattern in the sampled image includes: obtaining the target pattern corresponding to the measurement pattern in the sampled image; and obtaining the corresponding position of the measurement pattern in the sampled image by combining the position of the target pattern.

[0014] Optionally, obtaining the sampling contour of the measurement pattern at the corresponding position in multiple sampled images includes: obtaining the initial sampling contour of the measurement pattern in multiple sampled images; obtaining the size of the initial sampling contour; obtaining the actual size of the measurement pattern; calibrating the size of the initial sampling contour based on the actual size of the measurement pattern, and using the calibrated initial sampling contour as the sampling contour.

[0015] Optionally, based on the sampling contours of multiple sampled images, the graphic contour of the measurement graphic corresponding to the sampling position of the sampled image is obtained, including: obtaining the average contour of the sampling contours of multiple sampled images as the initial graphic contour; and smoothing the initial graphic contour to obtain the graphic contour.

[0016] Optionally, multiple measurement positions can be obtained on the graphic contour, including: filtering multiple initial measurement positions, removing initial measurement positions that are the same as those in the surrounding environment, and retaining the remaining initial measurement positions as measurement positions.

[0017] Optionally, filtering is performed on multiple initial measurement locations, including: grouping the spatial points corresponding to the initial measurement locations in multiple sampled images into the same coordinate system space; obtaining the spacing between the spatial points in the coordinate system space; when the spacing is less than or equal to the spacing threshold, removing either of the two corresponding spatial points and retaining the remaining spatial points; otherwise, retaining the corresponding spatial points.

[0018] Optionally, based on the graphic contour, the etching deviation value at the measurement position is obtained, including: obtaining the simulated exposure pattern corresponding to the measurement pattern; and obtaining the size difference between the graphic contour and the simulated exposure pattern at the measurement position as the etching deviation value.

[0019] Optionally, based on the etching deviation value at the measurement location, an etching deviation compensation model is established, including: obtaining the graphic feature value of the measurement location; and combining the etching deviation compensation value and the graphic feature value at the measurement location to obtain the etching deviation compensation model.

[0020] Optionally, by combining the etching deviation compensation value and the graphic feature value at the measurement position, an etching deviation compensation model is obtained, including: performing function fitting on the etching deviation compensation value and the graphic feature value at the measurement position, obtaining the coefficients corresponding to the graphic feature value, and the coefficients and the graphic feature value constitute the etching deviation compensation model.

[0021] Accordingly, this embodiment of the invention also provides a system for establishing an etching deviation compensation model, comprising: a measurement structure providing module for providing a measurement structure, including multiple measurement patterns, wherein the measurement patterns are etched patterns; a sampling position selection module for selecting sampling positions of the measurement structure, wherein a portion of the measurement pattern at the sampling position represents the feature information of the measurement structure; a sampling image acquisition module for acquiring multiple sampling images at the sampling positions; a pattern contour acquisition module for obtaining the pattern contour of the measurement pattern corresponding to the sampling position based on the multiple sampling images; a measurement position acquisition module for acquiring multiple measurement positions on the pattern contour; an etching deviation value acquisition module for obtaining the etching deviation value of the measurement position based on the pattern contour; and an etching deviation compensation model establishment module for establishing an etching deviation compensation model based on the etching deviation value of the measurement position.

[0022] Accordingly, embodiments of the present invention also provide an etching deviation compensation model, including an etching deviation compensation model obtained by using the etching deviation compensation model establishment method provided in embodiments of the present invention.

[0023] Accordingly, embodiments of the present invention also provide an etching deviation compensation method, including an etching deviation compensation method based on an etching deviation compensation model established by the etching deviation compensation model establishment method provided in embodiments of the present invention.

[0024] Accordingly, embodiments of the present invention also provide an apparatus including at least one memory and at least one processor, wherein the memory stores one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the etching deviation compensation method provided in the embodiments of the present invention.

[0025] Accordingly, embodiments of the present invention also provide a storage medium storing one or more computer instructions, which are used to implement the etching deviation compensation method provided in embodiments of the present invention.

[0026] Compared with the prior art, the technical solution of the embodiments of the present invention has the following advantages:

[0027] In the method for establishing an etching deviation compensation model provided in this embodiment of the invention, sampling positions of the measurement structure are selected, and partial measurement patterns at the sampling positions characterize the feature information of the measurement structure. Multiple sampling images of the sampling positions are obtained, and the graphic contours of the measurement patterns corresponding to the sampling positions are obtained based on the multiple sampling images. Multiple measurement positions are obtained on the graphic contours, and the etching deviation values ​​of the measurement positions are obtained based on the graphic contours. Based on the etching deviation values ​​of the measurement positions, an etching deviation compensation model is established. In this embodiment of the invention, selecting partial measurement patterns that can characterize the feature information of the measurement structure for establishing the etching deviation compensation model helps to reduce computing power, improve model establishment efficiency, and save computational costs. Furthermore, obtaining the graphic contours of the corresponding measurement patterns through multiple sampling images helps to improve computational accuracy, thereby improving the accuracy of the etching deviation compensation model. Attached Figure Description

[0028] Figure 1 This is a flowchart of an embodiment of the method for establishing the etching deviation compensation model of the present invention;

[0029] Figures 2 to 7 This is a schematic diagram of each step in one embodiment of the method for establishing the etching deviation compensation model of the present invention;

[0030] Figure 8 This is a functional block diagram of an embodiment of the etching deviation compensation model establishment system of the present invention;

[0031] Figure 9 This is a hardware structure diagram of an embodiment of the device provided by the present invention. Detailed Implementation

[0032] As the background technology indicates, currently, when establishing etching deviation compensation models, wafer data is typically collected through automated measurement using a scanning electron microscope (CD-SEM) to measure feature dimensions. However, automated CD-SEM measurement still suffers from drawbacks such as low data acquisition efficiency and inaccurate pattern position capture. Moreover, CD-SEM can only perform limited measurements from a single image, and two different images may contain the same environmental information, leading to high time and resource costs. Furthermore, the situation becomes even more severe if the pattern position is captured incorrectly.

[0033] Rule-based etching deviation compensation methods involve manually analyzing and comparing collected data to create an etching deviation table, enabling separate compensation for different types of patterns and structures. This manual process has limited accuracy and complexity, making it difficult to achieve satisfactory compensation results, especially as future designs become increasingly smaller.

[0034] Another approach is model-based etching deviation compensation. This typically requires defining appropriate model characteristic parameters to capture the loading effect of etching deviation, and has stricter requirements for data coverage than rule-based etching deviation compensation methods. Therefore, the Variable Etching Deviation (VEB) model requires a large amount of data, resulting in significant measurement time. However, CD-SEM measurements have limitations in efficiency and accuracy, which restricts the application of the VEB model in actual production.

[0035] To address the technical problem, embodiments of the present invention provide a method for establishing an etching deviation compensation model. (Reference) Figure 1 The flowchart illustrates an embodiment of the method for establishing the etching deviation compensation model of the present invention.

[0036] In this embodiment, the method for establishing the etching deviation compensation model includes the following basic steps:

[0037] Step S1: Provide a measurement structure, including multiple measurement patterns, which are etched patterns;

[0038] Step S2: Select the sampling location of the measurement structure. The partial measurement image of the sampling location represents the characteristic information of the measurement structure.

[0039] Step S3: Acquire multiple sampled images at the sampling locations;

[0040] Step S4: Obtain the graphic outline of the measurement pattern corresponding to the sampling position based on multiple sampled images;

[0041] Step S5: Obtain multiple measurement locations on the graphic contour;

[0042] Step S6: Obtain the etching deviation value at the measurement position based on the graphic contour;

[0043] Step S7: Establish an etching deviation compensation model based on the etching deviation value at the measurement position.

[0044] In the method for establishing an etching deviation compensation model provided in this embodiment of the invention, a portion of the measurement pattern that can characterize the feature information of the measurement structure is selected for establishing the etching deviation compensation model. This helps to reduce computing power, improve model establishment efficiency, and save computational costs. Furthermore, obtaining the graphic contour of the corresponding measurement pattern through multiple sampled images helps to improve computational accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0045] To make the above-mentioned objects, features and advantages of the embodiments of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Figures 2 to 7 This is a schematic diagram of each step in one embodiment of the method for establishing the etching deviation compensation model of the present invention.

[0047] refer to Figure 2 Step S1: Provide a measurement structure 100, including multiple measurement patterns 110, where the measurement patterns 110 are etched patterns.

[0048] In semiconductor manufacturing, the pattern on the mask is etched to form a corresponding structure on the wafer. The measurement structure 100 is a structural layer transferred from the mask layer to the wafer through the etching process. The measurement pattern 110 is a pattern structure transferred from the target pattern in the mask layer to the wafer through the etching process.

[0049] Specifically, in this embodiment, the measurement pattern 110 is used to subsequently obtain the etching deviation value.

[0050] Execution step S2: Select the sampling position of the measurement structure 100. The partial measurement pattern 110 of the sampling position represents the feature information of the measurement structure 100.

[0051] Among them, feature information refers to the information of the graphic features of the measuring structure 100, such as length, width, spacing, etc.

[0052] The sampling position is the position of the partial measurement pattern 110 selected in the measurement structure 100 that can characterize the feature information of the entire measurement structure 100. Subsequently, the etching deviation compensation model is established by using the measurement pattern 110 at the selected sampling position. This eliminates the need to perform comprehensive calculations on the entire measurement structure 100 and allows the selected sampling position to represent the feature information of the entire measurement structure 100, making the calculation accurate and efficient.

[0053] In this embodiment, a portion of the measurement pattern 110 that can characterize the feature information of the measurement structure 100 is selected to establish an etching deviation compensation model, which helps to reduce computing power, improve model establishment efficiency, and save computational costs.

[0054] In this embodiment, selecting the sampling position of the measurement structure 100 includes: obtaining multiple initial sampling positions of the measurement structure 100, wherein the measurement patterns 110 of the multiple initial sampling positions are different.

[0055] The initial sampling position covers almost the entire measurement structure 100, thus selecting an appropriate number of portions from the initial sampling position as sampling positions.

[0056] In this embodiment, the graphic feature information of the measurement graphic 110 at the initial sampling position is obtained.

[0057] Among them, graphic feature information refers to the graphic features of the measurement graphic 110, such as length, width, spacing, etc. The graphic feature information of the measurement graphic 110 in the measurement structure 100 can characterize the feature information of the measurement structure 100.

[0058] The graphic feature information of the measurement pattern 110 at the initial sampling position is obtained, and used to find the measurement pattern 110 corresponding to the graphic feature information that can characterize the feature information of the measurement structure 100.

[0059] In this embodiment, obtaining the graphic feature information of the measurement graphic 110 at the initial sampling position includes: obtaining the initial sampling image at the initial sampling position.

[0060] An initial sampling image of the initial sampling position is obtained, which is used to obtain the graphic feature information corresponding to the measurement graphic 110 of the initial sampling position based on the initial sampling image.

[0061] In this embodiment, multiple initial measurement positions (such as...) are set on the measurement pattern 110 of the initial sampled image. Figure 2 (As indicated by the double-headed arrow).

[0062] The initial measurement position is the specific location for subsequent measurements, such as obtaining the graphic dimensions, graphic feature values, and etching deviation values.

[0063] Specifically, in this embodiment, the coordinates (x, y) are used to represent the specific location of the initial measurement position.

[0064] In this embodiment, graphic feature values ​​of multiple initial measurement locations are obtained.

[0065] Among them, graphic feature values ​​refer to the specific numerical values ​​of the aforementioned graphic features, such as length dimension values, width dimension values, spacing values, etc.

[0066] The graphic feature values ​​of multiple initial measurement locations are obtained, which are then used to obtain the distribution of graphic feature information corresponding to the initial sampled image.

[0067] In this embodiment, obtaining the graphic feature value of the initial measurement position includes: setting model feature parameters based on the feature information of the measurement structure 100.

[0068] Among them, the model feature parameters are the feature information of the measurement structure 100 that is highly relevant to the establishment of the etching deviation compensation model, such as length, width, spacing, etc. In practical applications, appropriate model feature parameters are selected according to the actual situation of the current layer, such as density parameters containing short-range environmental information, interlayer parameters containing density information of the previous layer, etc.

[0069] Based on the feature information of the measurement structure 100, the model feature parameters are set for subsequent construction of the coordinate system space.

[0070] In this embodiment, the simulated exposure pattern corresponding to the measurement pattern 110 is obtained.

[0071] Among them, the simulated exposure pattern refers to the pattern that the measurement pattern 110 simulates when exposed in the photoresist.

[0072] The simulated exposure pattern corresponding to the measurement pattern 110 is obtained to obtain the value of the model feature parameters. Furthermore, the etching deviation value is obtained by the difference between the measurement pattern 110 and the simulated exposure pattern.

[0073] In this embodiment, the model feature parameters at the initial measurement position are obtained as graphic feature values ​​based on the simulated exposure pattern.

[0074] Specifically, in this embodiment, an initial measurement location corresponds to the values ​​of multiple model feature parameters.

[0075] In this embodiment, the simulated exposure pattern corresponding to the measurement pattern 110 is obtained from the initial sampled image based on the initial measurement position.

[0076] Specifically, in this embodiment, the simulated exposure pattern corresponding to the measurement pattern 110 is obtained according to the specific coordinates of the initial measurement position.

[0077] In this embodiment, the distribution of graphic feature information corresponding to each initial sampled image is obtained based on multiple graphic feature values.

[0078] By obtaining the distribution of graphic feature information corresponding to each initial sampled image, the set of graphic feature information distributions of multiple initial sampled images can be found to cover the feature information of the measurement structure 100.

[0079] refer to Figure 3 In this embodiment, the distribution of graphic feature information corresponding to each initial sampled image is obtained based on multiple graphic feature values, including: establishing a corresponding coordinate system space based on the model feature parameters, wherein the coordinate axes of the coordinate system space correspond one-to-one with the model feature parameters.

[0080] A corresponding coordinate system space is established based on the model's feature parameters, which is used to obtain the distribution of graphic feature information in the coordinate system space.

[0081] Specifically, as an example, such as Figure 3 As shown, this embodiment sets three model feature parameters: model feature parameter A, model feature parameter B, and model feature parameter C, which constitute a three-dimensional coordinate system space.

[0082] In this embodiment, the model feature parameters at the initial measurement position are used as the coordinates of the corresponding coordinate axes. Multiple initial measurement positions are set in the coordinate system space to obtain spatial points that correspond one-to-one with the initial measurement positions.

[0083] Specifically, as an example, such as Figure 3 As shown, the values ​​of the model feature parameters at the initial measurement position are the coordinates of the corresponding coordinate axes, that is, the coordinates (A, B, C) corresponding to each initial measurement position, where A, B, and C are the values ​​of model feature parameter A, model feature parameter B, and model feature parameter C corresponding to that initial measurement position, respectively.

[0084] refer to Figure 4 In this embodiment, based on the spatial point distribution of each initial sampled image, the coordinate system space of each initial sampled image is divided into multiple class regions, and the distribution of spatial points in multiple class regions is obtained as the distribution of graphic feature information.

[0085] In this embodiment, the K-means clustering method is used to divide the coordinate space of each initial sampled image into multiple class regions based on the spatial point distribution of each initial sampled image.

[0086] Specifically, the steps for using the K-means clustering method are as follows:

[0087] The first step is to determine a value of k, which is used to divide the coordinate space of each initial sampled image into K class regions.

[0088] As an example, such as Figure 4 As shown in A, the value of k in this embodiment is 3.

[0089] The second step is to randomly select k data points from the spatial points as centroids.

[0090] As an example, such as Figure 4 As shown in Figure A, three data points c1, c2, and c3 are randomly selected from the spatial points as the centroids.

[0091] The third step is to calculate the distance (e.g., Euclidean distance) between each point in the coordinate system space and each centroid. The point is assigned to the set to which the centroid it is closer to.

[0092] As an example, such as Figure 4 As shown in Figure B, the spatial points that are divided into sets corresponding to the centroids are connected by straight lines.

[0093] The fourth step is to group all the spatial points into sets, resulting in a total of k sets. Then, the centroid of each set is recalculated.

[0094] As an example, such as Figure 4 As shown in C, the positions of the centroids c1, c2, and c3 were recalculated based on the spatial points of each set.

[0095] Fifth step: If the distance between the newly calculated centroid and the original centroid is less than or equal to the set threshold (indicating that the position of the recalculated centroid has not changed much and tends to be stable, or in other words, converged), the clustering can be considered to have achieved the expected result, and the algorithm terminates.

[0096] As an example, such as Figure 4 As shown in D, spatial points are distributed in the divided class regions, obtaining the coordinate system spatial division of each initial sampled image.

[0097] It should be noted that if the distance between the newly calculated centroid and the original centroid changes significantly (i.e., exceeds the set threshold), then it is necessary to iterate through steps three through five.

[0098] In this embodiment, some initial sampling positions are selected as sampling positions, and the sum of the graphic feature information of some initial sampling positions covers the feature information of the measurement structure 100.

[0099] If the sum of the graphic feature information of a portion of the initial sampling positions covers the feature information of the measurement structure 100, then the measurement graphic 110 of that portion of the initial sampling positions can characterize the feature information of the measurement structure 100, and thus the initial sampling positions of that portion are selected as the sampling positions.

[0100] In this embodiment, some initial sampling positions are selected as sampling positions. Based on the distribution of feature information of the initial sampling images, the initial sampling positions corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measurement structure 100 are selected as sampling positions.

[0101] If the initial sampling positions corresponding to multiple initial sampling images that cover the feature information distribution of the measurement structure 100 are selected as sampling positions, then the measurement pattern 110 of the selected sampling positions can characterize the feature information of the measurement structure 100.

[0102] In this embodiment, based on the feature information distribution of the initial sampled images, the initial sampling positions corresponding to multiple initial sampled images whose feature information distribution covers the feature information of the measurement structure 100 are selected as sampling positions, including: obtaining the number of class regions covered by the spatial point distribution of each initial sampled image.

[0103] It should be noted that the coordinate system space of each initial sampled image is divided into k class regions, but ultimately each spatial point may be distributed in all class regions or in some class regions.

[0104] The number of class regions covered by the spatial point distribution of each initial sampling image can be obtained. Based on the coverage of multiple class regions by the spatial points of each initial sampling image, the feature information of the measurement structure 100 that each initial sampling image can represent can be obtained. Specifically, the more class regions covered by the spatial point distribution of the initial sampling image, the more feature information of the measurement structure 100 that the initial sampling image can represent.

[0105] In this embodiment, based on the number of class regions corresponding to the initial sampled images, multiple initial sampled images are selected in descending order of their number of initial sampled images as sampling positions. The distribution of the set of spatial points in the multiple initial sampled images covers all class regions.

[0106] Based on the number of class regions corresponding to the initial sampled images, multiple initial sampling positions corresponding to the initial sampled images are selected in descending order of their number of regions as sampling positions. In other words, based on the amount of feature information of the represented measurement structure 100, multiple initial sampling positions corresponding to the initial sampled images are selected in descending order of their number of regions as sampling positions.

[0107] It should be noted that in practical applications, the appropriate number of initial sampling graphics should be selected based on actual needs and computing power.

[0108] Step S3: Acquire multiple sampled images at the sampling locations.

[0109] Multiple sampling images of the sampling location are acquired to obtain the graphic outline of the measurement graphic 110 of the sampling location.

[0110] Specifically, multiple sampled images are acquired for each sampling location.

[0111] In this embodiment, CD-SEM is used to acquire multiple sampled images of the sampling location.

[0112] refer to Figure 5 Step S4: Obtain the graphic contour of the measurement graphic 110 corresponding to the sampling position based on multiple sampled images.

[0113] The graphic profile is used to subsequently obtain etching deviation values.

[0114] In this embodiment, obtaining the graphic contour of the corresponding measurement graphic 110 through multiple sampled images helps to improve the calculation accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0115] In this embodiment, obtaining the graphic outline of the measurement graphic 110 corresponding to the sampling position based on multiple sampling images includes: determining the corresponding position of the measurement graphic 110 in the sampling image.

[0116] Determining the corresponding position of the measurement pattern 110 in the sampled image helps improve the accuracy of obtaining the pattern contour.

[0117] In this embodiment, determining the corresponding position of the measurement pattern 110 in the sampled image includes: obtaining the target pattern corresponding to the measurement pattern 110 in the sampled image.

[0118] The target graphic corresponding to the measurement graphic 110 is used to calibrate the position of the measurement graphic 110.

[0119] Specifically, in this embodiment, the corresponding position of the measurement graphic 110 in the sampled image is obtained by combining the position of the target graphic.

[0120] In this embodiment, the sampling contours of the measurement pattern 110 at corresponding positions are obtained from multiple sampling images.

[0121] The sampling contours of the measurement pattern 110 at corresponding positions in multiple sampled images are obtained, which are used to obtain the graphic contours of the measurement pattern 110 at the sampling positions corresponding to the sampled images.

[0122] As an example, Figure 5 (a) shows multiple sampling profiles corresponding to the acquired measurement pattern 110.

[0123] In this embodiment, obtaining the sampling contour of the measurement pattern 110 at the corresponding position in multiple sampling images includes: obtaining the initial sampling contour of the measurement pattern 110 in multiple sampling images; obtaining the size of the initial sampling contour; obtaining the actual size of the measurement pattern 110; calibrating the size of the initial sampling contour based on the actual size of the measurement pattern 110; and using the calibrated initial sampling contour as the sampling contour.

[0124] Due to image bias, the initial sampling profile obtained in the sampled image will deviate from the actual size of the measured pattern 110. Therefore, the size of the initial sampling profile is calibrated with the actual size of the measured pattern 110 to obtain a more accurate sampling profile.

[0125] In this embodiment, the size of the initial sampling profile is adjusted to be the same as the actual size, based on the actual size of the measured pattern 110, to complete the calibration of the initial sampling profile and obtain the sampling profile.

[0126] In this embodiment, CD-SEM is used to obtain the initial sampling contour of the measurement pattern 110 at the corresponding position.

[0127] In this embodiment, CD-SEM is used to obtain the actual dimensions of the measurement pattern 110.

[0128] In this embodiment, the graphic contour of the measurement graphic 110 corresponding to the sampling position of the sampling image is obtained based on the sampling contour of multiple sampling images.

[0129] For a given sampling location, the graphic contour of the measurement graphic 110 corresponding to the sampling location is obtained based on the sampling contours of multiple sampling images, which helps to improve the accuracy of graphic contour acquisition.

[0130] In this embodiment, the graphic contour of the measurement graphic 110 corresponding to the sampling position of the sampling image is obtained based on the sampling contours of multiple sampling images, including: obtaining the average contour of the sampling contours of multiple sampling images as the initial graphic contour.

[0131] The average contour of the sampled contours of multiple sampled images is obtained as the initial graphic contour, which can take into account the results of multiple sampled images and improve the accuracy of the initial graphic contour.

[0132] In this embodiment, the initial graphic outline is smoothed to obtain the graphic outline.

[0133] The initial graphic contour is smoothed to remove some abrupt errors and further improve the accuracy of the graphic contour.

[0134] Perform step S5: Obtain multiple measurement locations on the graphic contour.

[0135] The measurement location is used as the location to obtain the etching deviation value.

[0136] In this embodiment, obtaining multiple measurement positions on the graphic contour includes: filtering multiple initial measurement positions, removing initial measurement positions with the same surrounding environment, and retaining the remaining initial measurement positions as measurement positions.

[0137] For initial measurement positions with the same surrounding environment, there is no need to repeat the measurement. Therefore, in this embodiment, multiple initial measurement positions are filtered to remove those with the same surrounding environment and retain the remaining initial measurement positions as measurement positions. This helps to improve the computational efficiency of obtaining etching deviation values ​​at the measurement positions.

[0138] In this embodiment, filtering is performed on multiple initial measurement positions, including: grouping the spatial points corresponding to the initial measurement positions in multiple sampled images into the same coordinate system space; obtaining the spacing between the spatial points in the coordinate system space; when the spacing is less than or equal to the spacing threshold, removing either of the two corresponding spatial points and retaining the remaining spatial points; otherwise, retaining the corresponding spatial points.

[0139] When the spacing between spatial points in the coordinate system space is small (i.e., less than or equal to the spacing threshold), it indicates that the environments around the initial measurement positions represented by the two spatial points are similar, so one of the two corresponding spatial points needs to be removed. When the spacing between spatial points in the coordinate system space is large (i.e., greater than the spacing threshold), it indicates that the environments around the initial measurement positions represented by the two spatial points are different, so the corresponding spatial points need to be retained.

[0140] In this embodiment, a corresponding spacing threshold is set according to actual process requirements.

[0141] As an example, Figure 6 The coordinate system space before filtering is shown. Figure 7 The coordinate system space after filtering is shown.

[0142] Step S6: Obtain the etching deviation value of the measurement position based on the graphic contour.

[0143] Obtain the etching deviation value at the measurement location, which will be used as the output value for the subsequent establishment of the etching deviation compensation model.

[0144] In this embodiment, obtaining the etching deviation value at the measurement position based on the graphic contour includes: obtaining the size difference between the graphic contour and the simulated exposure graphic at the measurement position as the etching deviation value.

[0145] The size difference between the pattern outline and the simulated exposure pattern is the deviation of the structure obtained after the actual etching process due to the etching process. Therefore, the size difference between the pattern outline and the simulated exposure pattern at the measurement location is used as the etching deviation value.

[0146] Step S7: Establish an etching deviation compensation model based on the etching deviation value at the measurement position.

[0147] Based on the etching deviation value at the measurement position, the etching deviation compensation model is obtained through continuous iteration.

[0148] In this embodiment, an etching deviation compensation model is established based on the etching deviation value at the measurement location, including: obtaining the graphic feature value of the measurement location.

[0149] The graphic feature value of the measurement location is the graphic feature value of the initial measurement location selected above. After filtering, the measurement location is obtained, and the corresponding graphic feature value of the measurement location is obtained.

[0150] In this embodiment, the etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement position.

[0151] The etching deviation compensation model is established by using the etching deviation compensation value at the measurement location as the output value and the graphic feature value at the measurement location as the input value.

[0152] In this embodiment, the etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement position. This includes: performing function fitting on the etching deviation compensation value and the graphic feature value at the measurement position to obtain the coefficients corresponding to the graphic feature values. The coefficients and the graphic feature values ​​constitute the etching deviation compensation model.

[0153] Different graphic feature values ​​are set as different polynomials. The etching deviation compensation value and graphic feature value at the measurement position are fitted by a function to obtain the coefficients of the polynomial corresponding to each graphic feature value. These coefficients are the calculation coefficients of the etching deviation compensation model. After the etching deviation compensation model is established, the graphic feature value of the position to be calculated is input to obtain the corresponding etching deviation compensation value.

[0154] Accordingly, the present invention also provides a system for establishing an etching deviation compensation model. Figure 8 This is a functional block diagram of an embodiment of the etching deviation compensation model establishment system of the present invention.

[0155] In this embodiment, the etching deviation compensation model establishment system 50 includes: a measurement structure providing module 501, used to provide a measurement structure, including multiple measurement patterns, wherein the measurement patterns are etched patterns; a sampling position selection module 502, used to select sampling positions of the measurement structure, wherein a portion of the measurement pattern at the sampling position represents the feature information of the measurement structure; a sampling image acquisition module 503, used to acquire multiple sampling images at the sampling positions; a pattern contour acquisition module 504, used to obtain the pattern contour of the measurement pattern corresponding to the sampling position based on the multiple sampling images; a measurement position acquisition module 505, used to acquire multiple measurement positions on the pattern contour; an etching deviation value acquisition module 506, used to acquire the etching deviation value of the measurement position based on the pattern contour; and an etching deviation compensation model establishment module 507, used to establish an etching deviation compensation model based on the etching deviation value of the measurement position.

[0156] The measurement structure providing module 501 is used to provide a measurement structure, including multiple measurement patterns, which are etched patterns.

[0157] In semiconductor manufacturing, the pattern on the mask is etched to form a corresponding structure on the wafer. The measurement structure is the structural layer transferred from the mask layer to the wafer through the etching process. The measurement pattern is the target pattern in the mask layer transferred to the wafer through the etching process.

[0158] Specifically, in this embodiment, the measurement pattern is used to subsequently obtain the etching deviation value.

[0159] The sampling position selection module 502 is used to select the sampling position of the measurement structure, and the partial measurement graphic of the sampling position represents the characteristic information of the measurement structure.

[0160] Among them, feature information refers to the graphic features of the measured structure, such as length, width, spacing, etc.

[0161] The sampling position is the location of a portion of the measurement pattern selected in the measurement structure that can represent the feature information of the entire measurement structure. Subsequently, the etching deviation compensation model is established by using the measurement pattern of the selected sampling position. This eliminates the need to perform comprehensive calculations on the entire measurement structure and allows the selected sampling position to represent the feature information of the entire measurement structure, making the calculation accurate and efficient.

[0162] In this embodiment, partial measurement patterns that can characterize the measurement structure are selected to establish an etching deviation compensation model, which helps to reduce computing power, improve model establishment efficiency, and save computational costs.

[0163] In this embodiment, selecting the sampling position of the measurement structure includes: obtaining multiple initial sampling positions of the measurement structure, wherein the measurement patterns of the multiple initial sampling positions are different.

[0164] The initial sampling location covers almost the entire measurement structure, thus allowing an appropriate number of portions to be selected as sampling locations from the initial sampling location.

[0165] In this embodiment, graphic feature information of the measurement pattern at the initial sampling position is obtained.

[0166] Among them, graphic feature information refers to the graphic features of the measured graphic, such as length, width, spacing, etc. The graphic feature information of the measured graphic in the measurement structure can characterize the feature information of the measurement structure.

[0167] The graphic feature information of the measurement pattern at the initial sampling position is obtained, which is used to find the measurement pattern corresponding to the graphic feature information that can characterize the measurement structure.

[0168] In this embodiment, obtaining the graphic feature information of the measurement pattern at the initial sampling position includes: obtaining the initial sampling image at the initial sampling position.

[0169] The initial sampling image at the initial sampling position is obtained, which is used to obtain the graphic feature information corresponding to the measurement graphic at the initial sampling position based on the initial sampling image.

[0170] In this embodiment, multiple initial measurement positions are set on the measurement pattern of the initial sampled image.

[0171] The initial measurement position is the specific location for subsequent measurements, such as obtaining the graphic dimensions, graphic feature values, and etching deviation values.

[0172] Specifically, in this embodiment, the coordinates (x, y) are used to represent the specific location of the initial measurement position.

[0173] In this embodiment, graphic feature values ​​of multiple initial measurement locations are obtained.

[0174] Among them, graphic feature values ​​refer to the specific numerical values ​​of the aforementioned graphic features, such as length dimension values, width dimension values, spacing values, etc.

[0175] The graphic feature values ​​of multiple initial measurement locations are obtained, which are then used to obtain the distribution of graphic feature information corresponding to the initial sampled image.

[0176] In this embodiment, obtaining the graphic feature values ​​of the initial measurement position includes: setting model feature parameters based on the feature information of the measurement structure.

[0177] Among them, the model feature parameters are the feature information of the measurement structure that is highly relevant to the establishment of the etching deviation compensation model, such as length, width, spacing, etc. In practical applications, appropriate model feature parameters are selected according to the actual situation of the current layer, such as density parameters that include short-range environmental information, interlayer parameters that include density information of the previous layer, etc.

[0178] Based on the characteristic information of the measured structure, the model characteristic parameters are set for subsequent construction of the coordinate system space.

[0179] In this embodiment, the simulated exposure pattern corresponding to the measured pattern is obtained.

[0180] Among them, simulated exposure pattern refers to the pattern that is measured to simulate exposure in photoresist.

[0181] The simulated exposure pattern corresponding to the measured pattern is obtained to obtain the value of the model feature parameters. Furthermore, the etching deviation value is obtained by the difference between the measured pattern and the simulated exposure pattern.

[0182] In this embodiment, the model feature parameters at the initial measurement position are obtained as graphic feature values ​​based on the simulated exposure pattern.

[0183] Specifically, in this embodiment, an initial measurement location corresponds to the values ​​of multiple model feature parameters.

[0184] In this embodiment, the simulated exposure pattern corresponding to the measured pattern is obtained from the initial sampled image based on the initial measurement position.

[0185] Specifically, in this embodiment, the simulated exposure pattern corresponding to the measurement pattern is obtained based on the specific coordinates of the initial measurement position.

[0186] In this embodiment, the distribution of graphic feature information corresponding to each initial sampled image is obtained based on multiple graphic feature values.

[0187] By obtaining the distribution of graphic feature information corresponding to each initial sampled image, the set of graphic feature information distributions of multiple initial sampled images can be found to cover the feature information of the measurement structure.

[0188] In this embodiment, the distribution of graphic feature information corresponding to each initial sampled image is obtained based on multiple graphic feature values, including: establishing a corresponding coordinate system space based on the model feature parameters, wherein the coordinate axes of the coordinate system space correspond one-to-one with the model feature parameters.

[0189] A corresponding coordinate system space is established based on the model's feature parameters, which is used to obtain the distribution of graphic feature information in the coordinate system space.

[0190] Specifically, as an example, such as Figure 3 As shown, this embodiment sets three model feature parameters: model feature parameter A, model feature parameter B, and model feature parameter C, which constitute a three-dimensional coordinate system space.

[0191] In this embodiment, the model feature parameters at the initial measurement position are used as the coordinates of the corresponding coordinate axes. Multiple initial measurement positions are set in the coordinate system space to obtain spatial points that correspond one-to-one with the initial measurement positions.

[0192] Specifically, as an example, such as Figure 3 As shown, the values ​​of the model feature parameters at the initial measurement position are the coordinates of the corresponding coordinate axes, that is, the coordinates (A, B, C) corresponding to each initial measurement position, where A, B, and C are the values ​​of model feature parameter A, model feature parameter B, and model feature parameter C corresponding to that initial measurement position, respectively.

[0193] In this embodiment, based on the spatial point distribution of each initial sampled image, the coordinate system space of each initial sampled image is divided into multiple class regions, and the distribution of spatial points in multiple class regions is obtained as the distribution of graphic feature information.

[0194] In this embodiment, the K-means clustering method is used to divide the coordinate space of each initial sampled image into multiple class regions based on the spatial point distribution of each initial sampled image.

[0195] Specifically, the steps for using the K-means clustering method are as follows:

[0196] The first step is to determine a value of k, which is used to divide the coordinate space of each initial sampled image into K class regions.

[0197] As an example, such as Figure 4 As shown in A, the value of k in this embodiment is 3.

[0198] The second step is to randomly select k data points from the spatial points as centroids.

[0199] As an example, such as Figure 4 As shown in Figure A, three data points c1, c2, and c3 are randomly selected from the spatial points as the centroids.

[0200] The third step is to calculate the distance (e.g., Euclidean distance) between each point in the coordinate system space and each centroid. The point is assigned to the set to which the centroid it is closer to.

[0201] As an example, such as Figure 4 As shown in Figure B, the spatial points that are divided into sets corresponding to the centroids are connected by straight lines.

[0202] The fourth step is to group all the spatial points into sets, resulting in a total of k sets. Then, the centroid of each set is recalculated.

[0203] As an example, such as Figure 4 As shown in C, the positions of the centroids c1, c2, and c3 were recalculated based on the spatial points of each set.

[0204] Fifth step: If the distance between the newly calculated centroid and the original centroid is less than or equal to the set threshold (indicating that the position of the recalculated centroid has not changed much and tends to be stable, or in other words, converged), the clustering can be considered to have achieved the expected result, and the algorithm terminates.

[0205] As an example, such as Figure 4 As shown in D, spatial points are distributed in the divided class regions, obtaining the coordinate system spatial division of each initial sampled image.

[0206] It should be noted that if the distance between the newly calculated centroid and the original centroid changes significantly (i.e., exceeds the set threshold), then it is necessary to iterate through steps three through five.

[0207] In this embodiment, a portion of the initial sampling positions are selected as sampling positions, and the sum of the graphic feature information of the portion of the initial sampling positions covers the feature information of the measured structure.

[0208] If the sum of the graphic feature information of a portion of the initial sampling positions covers the feature information of the measured structure, then the measurement graphic of that portion of the initial sampling positions can characterize the feature information of the measured structure, and thus that portion of the initial sampling positions is selected as the sampling positions.

[0209] In this embodiment, some initial sampling positions are selected as sampling positions. Based on the distribution of feature information of the initial sampling images, the initial sampling positions corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measurement structure are selected as sampling positions.

[0210] If the initial sampling positions corresponding to multiple initial sampling images that cover the feature information distribution of the measurement structure are selected as the sampling positions, then the measurement pattern of the selected sampling positions can characterize the feature information of the measurement structure.

[0211] In this embodiment, based on the feature information distribution of the initial sampled images, the initial sampling positions corresponding to multiple initial sampled images whose feature information distribution covers the feature information of the measurement structure are selected as sampling positions, including: obtaining the number of class regions covered by the spatial point distribution of each initial sampled image.

[0212] It should be noted that the coordinate system space of each initial sampled image is divided into k class regions, but ultimately each spatial point may be distributed in all class regions or in some class regions.

[0213] The number of class regions covered by the spatial point distribution of each initial sampled image can be obtained. Based on the coverage of multiple class regions by the spatial points of each initial sampled image, the feature information of the measurement structure that each initial sampled image can represent can be obtained. Specifically, the more class regions covered by the spatial point distribution of the initial sampled image, the more feature information of the measurement structure that the initial sampled image can represent.

[0214] In this embodiment, based on the number of class regions corresponding to the initial sampled images, multiple initial sampled images are selected in descending order of their number of initial sampled images as sampling positions. The distribution of the set of spatial points in the multiple initial sampled images covers all class regions.

[0215] Based on the number of class regions corresponding to the initial sampled images, multiple initial sampling positions corresponding to the initial sampled images are selected in descending order of their number of regions as sampling positions. In other words, based on the amount of feature information of the represented measurement structure, multiple initial sampling positions corresponding to the initial sampled images are selected in descending order of their number of regions as sampling positions.

[0216] It should be noted that in practical applications, the appropriate number of initial sampling graphics should be selected based on actual needs and computing power.

[0217] The sampling image acquisition module 503 is used to acquire multiple sampling images at the sampling location.

[0218] Multiple sampled images of the sampling location are acquired to obtain the graphic outline of the measurement pattern at the sampling location.

[0219] Specifically, multiple sampled images are acquired for each sampling location.

[0220] In this embodiment, CD-SEM is used to acquire multiple sampled images of the sampling location.

[0221] The graphic contour acquisition module 504 is used to obtain the graphic contour of the measurement graphic corresponding to the sampling position based on multiple sampled images.

[0222] The graphic profile is used to subsequently obtain etching deviation values.

[0223] In this embodiment, obtaining the graphic contour of the corresponding measurement pattern through multiple sampled images helps to improve the calculation accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0224] In this embodiment, obtaining the graphic outline of the measurement graphic corresponding to the sampling position based on multiple sampling images includes: determining the corresponding position of the measurement graphic in the sampling image.

[0225] Determining the corresponding position of the measured shape in the sampled image helps improve the accuracy of obtaining the shape contour.

[0226] In this embodiment, determining the corresponding position of the measurement graphic in the sampled image includes: obtaining the target graphic corresponding to the measurement graphic in the sampled image.

[0227] The target graphic corresponding to the measured graphic is used to calibrate the position of the measured graphic.

[0228] Specifically, in this embodiment, the corresponding position of the measured graphic in the sampled image is obtained by combining the position of the target graphic.

[0229] In this embodiment, the sampling contours of the measured pattern at corresponding positions are obtained from multiple sampled images.

[0230] The sample contours of the measured shapes at corresponding positions in multiple sampled images are obtained, which are used to obtain the shape contours of the measured shapes at the corresponding sampling positions in the sampled images.

[0231] In this embodiment, obtaining the sampling contour of the measurement pattern at the corresponding position in multiple sampling images includes: obtaining the initial sampling contour of the measurement pattern in multiple sampling images; obtaining the size of the initial sampling contour; obtaining the actual size of the measurement pattern; calibrating the size of the initial sampling contour based on the actual size of the measurement pattern; and using the calibrated initial sampling contour as the sampling contour.

[0232] Due to image bias, the initial sampling contour obtained from the sampled image will deviate from the actual size of the measured image. Therefore, the size of the initial sampling contour is calibrated with the actual size of the measured image to obtain a more accurate sampling contour.

[0233] In this embodiment, the size of the initial sampling contour is adjusted to be the same as the actual size, based on the actual size of the measured graphic, to complete the calibration of the initial sampling contour and obtain the sampling contour.

[0234] In this embodiment, CD-SEM is used to obtain the initial sampling contour of the measurement pattern at the corresponding position.

[0235] In this embodiment, CD-SEM is used to obtain the actual dimensions of the measured graphic.

[0236] In this embodiment, the graphic contour of the measurement graphic corresponding to the sampling position of the sampling image is obtained based on the sampling contours of multiple sampling images.

[0237] For a given sampling location, the graphic contour of the measured graphic at the corresponding sampling location can be obtained based on the sampling contours of multiple sampling images, which helps to improve the accuracy of graphic contour acquisition.

[0238] In this embodiment, the graphic contour of the measurement graphic corresponding to the sampling position of the sampling image is obtained based on the sampling contours of multiple sampling images, including: obtaining the average contour of the sampling contours of multiple sampling images as the initial graphic contour.

[0239] The average contour of the sampled contours of multiple sampled images is obtained as the initial graphic contour, which can take into account the results of multiple sampled images and improve the accuracy of the initial graphic contour.

[0240] In this embodiment, the initial graphic outline is smoothed to obtain the graphic outline.

[0241] The initial graphic contour is smoothed to remove some abrupt errors and further improve the accuracy of the graphic contour.

[0242] The measurement position acquisition module 505 is used to acquire multiple measurement positions on the graphic contour.

[0243] The measurement location is used as the location to obtain the etching deviation value.

[0244] In this embodiment, obtaining multiple measurement positions on the graphic contour includes: filtering multiple initial measurement positions, removing initial measurement positions with the same surrounding environment, and retaining the remaining initial measurement positions as measurement positions.

[0245] For initial measurement positions with the same surrounding environment, there is no need to repeat the measurement. Therefore, in this embodiment, multiple initial measurement positions are filtered to remove those with the same surrounding environment and retain the remaining initial measurement positions as measurement positions. This helps to improve the computational efficiency of obtaining etching deviation values ​​at the measurement positions.

[0246] In this embodiment, filtering is performed on multiple initial measurement positions, including: grouping the spatial points corresponding to the initial measurement positions in multiple sampled images into the same coordinate system space; obtaining the spacing between the spatial points in the coordinate system space; when the spacing is less than or equal to the spacing threshold, removing either of the two corresponding spatial points and retaining the remaining spatial points; otherwise, retaining the corresponding spatial points.

[0247] When the spacing between spatial points in the coordinate system space is small (i.e., less than or equal to the spacing threshold), it indicates that the environments around the initial measurement positions represented by the two spatial points are similar, so one of the two corresponding spatial points needs to be removed. When the spacing between spatial points in the coordinate system space is large (i.e., greater than the spacing threshold), it indicates that the environments around the initial measurement positions represented by the two spatial points are different, so the corresponding spatial points need to be retained.

[0248] In this embodiment, a corresponding spacing threshold is set according to actual process requirements.

[0249] As an example, Figure 6 The coordinate system space before filtering is shown. Figure 7 The coordinate system space after filtering is shown.

[0250] The etching deviation value acquisition module 506 is used to acquire the etching deviation value of the measurement position based on the graphic contour.

[0251] Obtain the etching deviation value at the measurement location, which will be used as the output value for the subsequent establishment of the etching deviation compensation model.

[0252] In this embodiment, obtaining the etching deviation value at the measurement position based on the graphic contour includes: obtaining the size difference between the graphic contour and the simulated exposure graphic at the measurement position as the etching deviation value.

[0253] The size difference between the pattern outline and the simulated exposure pattern is the deviation of the structure obtained after the actual etching process due to the etching process. Therefore, the size difference between the pattern outline and the simulated exposure pattern at the measurement location is used as the etching deviation value.

[0254] The etching deviation compensation model establishment module 507 is used to establish an etching deviation compensation model based on the etching deviation value of the measurement position.

[0255] Based on the etching deviation value at the measurement position, the etching deviation compensation model is obtained through continuous iteration.

[0256] In this embodiment, an etching deviation compensation model is established based on the etching deviation value at the measurement location, including: obtaining the graphic feature value of the measurement location.

[0257] The graphic feature value of the measurement location is the graphic feature value of the initial measurement location selected above. After filtering, the measurement location is obtained, and the corresponding graphic feature value of the measurement location is obtained.

[0258] In this embodiment, the etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement position.

[0259] The etching deviation compensation model is established by using the etching deviation compensation value at the measurement location as the output value and the graphic feature value at the measurement location as the input value.

[0260] In this embodiment, the etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement position. This includes: performing function fitting on the etching deviation compensation value and the graphic feature value at the measurement position to obtain the coefficients corresponding to the graphic feature values. The coefficients and the graphic feature values ​​constitute the etching deviation compensation model.

[0261] Different graphic feature values ​​are set as different polynomials. The etching deviation compensation value and graphic feature value at the measurement position are fitted by a function to obtain the coefficients of the polynomial corresponding to each graphic feature value. These coefficients are the calculation coefficients of the etching deviation compensation model. After the etching deviation compensation model is established, the graphic feature value of the position to be calculated is input to obtain the corresponding etching deviation compensation value.

[0262] Accordingly, embodiments of the present invention also provide an etching deviation compensation model, including an etching deviation compensation model obtained by using the etching deviation compensation model establishment method provided in embodiments of the present invention.

[0263] As can be seen from the foregoing embodiments, in the embodiments of the present invention, selecting a portion of the measurement pattern that can characterize the feature information of the measurement structure to establish the etching deviation compensation model is beneficial to reducing computing power, improving model establishment efficiency, and saving computational costs. Furthermore, obtaining the graphic contour of the corresponding measurement pattern through multiple sampled images is beneficial to improving computational accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0264] Accordingly, embodiments of the present invention also provide an etching deviation compensation method, including an etching deviation compensation method based on an etching deviation compensation model established by the etching deviation compensation model establishment method provided in embodiments of the present invention.

[0265] As can be seen from the foregoing embodiments, in the embodiments of the present invention, selecting a portion of the measurement pattern that can characterize the feature information of the measurement structure to establish the etching deviation compensation model is beneficial to reducing computing power, improving model establishment efficiency, and saving computational costs. Furthermore, obtaining the graphic contour of the corresponding measurement pattern through multiple sampled images is beneficial to improving computational accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0266] This invention also provides a device that can implement the etching deviation compensation model establishment method provided in this invention through the above-described etching deviation compensation model establishment method in the form of a loading program. An optional hardware structure of the terminal device provided in this invention can be as follows: Figure 9 As shown, it includes: at least one processor 01, at least one communication interface 02, at least one memory 03, and at least one communication bus 04.

[0267] In this embodiment, the number of processor 01, communication interface 02, memory 03, and communication bus 04 is at least one, and the processor 01, communication interface 02, and memory 03 communicate with each other through communication bus 04. Communication interface 02 can be an interface of a communication module used for network communication, such as the interface of a GSM module. Processor 01 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. Memory 03 may include high-speed RAM and may also include non-volatile memory (NVM), such as at least one disk storage device. Memory 03 stores one or more computer instructions, which are executed by processor 01 to implement the etching deviation compensation model establishment method provided in this embodiment of the present invention.

[0268] It should be noted that the aforementioned terminal device may also include other devices (not shown) that may not be essential to understanding the content disclosed in the embodiments of the present invention; given that these other devices may not be essential for understanding the content disclosed in the embodiments of the present invention, the embodiments of the present invention will not describe them one by one.

[0269] This invention also provides a storage medium storing one or more computer instructions for implementing the etching deviation compensation model establishment method provided in this invention.

[0270] In this embodiment of the invention, selecting a portion of the measurement pattern that can characterize the measurement structure to establish an etching deviation compensation model is beneficial for reducing computing power, improving model establishment efficiency, and saving computational costs. Furthermore, obtaining the graphic contour of the corresponding measurement pattern through multiple sampled images is beneficial for improving computational accuracy, thereby improving the accuracy of the etching deviation compensation model.

[0271] The embodiments of the present invention described above are combinations of elements and features of the present invention. Unless otherwise stated, elements or features may be considered optional. Individual elements or features may be practiced without combination with other elements or features. Furthermore, embodiments of the present invention may be constructed by combining some elements and / or features. The order of operations described in the embodiments of the present invention may be rearranged. Some constructions of any embodiment may be included in another embodiment and may be replaced by corresponding constructions of another embodiment. It will be apparent to those skilled in the art that claims in the appended claims that are not expressly referenced in each other may be combined to form embodiments of the present invention, or may be included as new claims in amendments made after the filing of this application.

[0272] Embodiments of the present invention can be implemented by various means, such as hardware, firmware, software, or combinations thereof. In a hardware configuration, the method according to an exemplary embodiment of the present invention can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, etc. In a firmware or software configuration, embodiments of the present invention can be implemented in the form of modules, processes, functions, etc. Software code can be stored in memory units and executed by a processor. The memory units are located inside or outside the processor and can send data to and receive data from the processor via various known means.

[0273] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is accorded the widest scope consistent with the principles and novel features disclosed herein.

[0274] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A method for establishing an etching deviation compensation model, characterized in that, include: A measurement structure is provided, including multiple measurement patterns, wherein the measurement patterns are etched patterns; A sampling location of the measurement structure is selected, and a partial measurement image of the sampling location represents the feature information of the measurement structure. Selecting a portion of the initial sampling locations as sampling locations, the sum of the graphic feature information of the portion of the initial sampling locations covers the feature information of the measured structure; Among the initial sampling locations selected as sampling locations, based on the distribution of feature information in the initial sampling images, the initial sampling locations corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measured structure are selected as sampling locations. Acquire multiple sampled images at the sampling location; The graphic outline of the measurement pattern corresponding to the sampling position is obtained from multiple sampled images; Multiple measurement positions are obtained on the graphic contour; Based on the graphic contour, the etching deviation value at the measurement position is obtained; Based on the etching deviation value at the measurement location, an etching deviation compensation model is established.

2. The method for establishing the etching deviation compensation model as described in claim 1, characterized in that, Selecting the sampling position of the measurement structure includes: obtaining multiple initial sampling positions of the measurement structure, wherein the measurement patterns of the multiple initial sampling positions are different; Obtain graphic feature information of the measurement pattern at the initial sampling position; A portion of the initial sampling positions are selected as the sampling positions, and the sum of the graphic feature information of the portion of the initial sampling positions covers the feature information of the measurement structure.

3. The method for establishing the etching deviation compensation model as described in claim 2, characterized in that, Obtaining graphic feature information of the measurement image at the initial sampling position includes: obtaining the initial sampling image at the initial sampling position; Multiple initial measurement positions are set on the measurement pattern of the initial sampled image; Obtain graphic feature values ​​for multiple initial measurement locations; Based on the multiple graphic feature values, the distribution of graphic feature information corresponding to each initial sampled image is obtained; Selecting a portion of the initial sampling positions as the sampling positions, and based on the feature information distribution of the initial sampling images, selecting the initial sampling positions corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measurement structure, as the sampling positions.

4. The method for establishing the etching deviation compensation model as described in claim 3, characterized in that, Obtaining the graphic feature values ​​of the initial measurement position includes: setting model feature parameters based on the feature information of the measurement structure; Obtain the simulated exposure pattern corresponding to the measured pattern; Based on the simulated exposure pattern, the value of the model feature parameter at the initial measurement position is obtained as the pattern feature value.

5. The method for establishing the etching deviation compensation model as described in claim 4, characterized in that, In obtaining the simulated exposure pattern corresponding to the measured pattern, based on the initial measurement position, the simulated exposure pattern corresponding to the measured pattern is obtained in the initial sampled image.

6. The method for establishing the etching deviation compensation model as described in claim 4, characterized in that, Based on multiple graphic feature values, the distribution of graphic feature information corresponding to each initial sampled image is obtained, including: establishing a corresponding coordinate system space based on model feature parameters, wherein the coordinate axes of the coordinate system space correspond one-to-one with the model feature parameters; Using the model feature parameters at the initial measurement position as the coordinates of the corresponding coordinate axes, multiple initial measurement positions are set in the coordinate system space to obtain spatial points that correspond one-to-one with the initial measurement positions; Based on the spatial point distribution of each initial sampled image, the coordinate system space of each initial sampled image is divided into multiple class regions, and the distribution of the spatial points in the multiple class regions is obtained as the distribution of the graphic feature information.

7. The method for establishing the etching deviation compensation model as described in claim 6, characterized in that, The K-means clustering method is used to divide the coordinate space of each initial sampled image into multiple class regions based on the spatial point distribution of each initial sampled image.

8. The method for establishing the etching deviation compensation model as described in claim 6, characterized in that, Based on the feature information distribution of the initial sampled images, select the initial sampling positions corresponding to multiple initial sampled images whose feature information distribution covers the feature information of the measurement structure, as the sampling positions, including: obtaining the number of class regions covered by the spatial point distribution of each initial sampled image; Based on the number of class regions corresponding to the initial sampled image, multiple initial sampled images are selected in descending order of their number of initial sampled positions as the sampling positions, wherein the set of spatial points in the multiple initial sampled images covers all the class regions.

9. The method for establishing the etching deviation compensation model as described in claim 1, characterized in that, Obtaining the graphic outline of the measurement graphic corresponding to the sampling position based on multiple sampling images includes: determining the corresponding position of the measurement graphic in the sampling image; Obtain the sampling contour of the measured pattern at corresponding positions in multiple sampled images; Based on the sampling contours of multiple sampled images, obtain the graphic contour of the measurement graphic at the sampling position corresponding to the sampled image.

10. The method for establishing the etching deviation compensation model as described in claim 9, characterized in that, Determining the corresponding position of the measurement pattern in the sampled image includes: obtaining the target pattern corresponding to the measurement pattern in the sampled image; Based on the position of the target graphic, the corresponding position of the measured graphic in the sampled image is obtained.

11. The method for establishing the etching deviation compensation model as described in claim 9, characterized in that, Obtaining the sampling contour of the measurement pattern at a corresponding position in multiple sampling images includes: obtaining the initial sampling contour of the measurement pattern in multiple sampling images; Obtain the dimensions of the initial sampled contour; Obtain the actual dimensions of the measured graphic; Using the actual size of the measured graphic as a reference, the size of the initial sampling contour is calibrated, and the calibrated initial sampling contour is used as the sampling contour.

12. The method for establishing the etching deviation compensation model as described in claim 9, characterized in that, Based on the sampling contours of multiple sampling images, the graphic contour of the measured graphic at the sampling position corresponding to the sampling image is obtained, including: obtaining the average contour of the sampling contours of multiple sampling images as the initial graphic contour; The initial graphic outline is smoothed to obtain the graphic outline.

13. The method for establishing the etching deviation compensation model as described in claim 6, characterized in that, Obtaining multiple measurement positions on the graphic contour includes: filtering the multiple initial measurement positions, removing initial measurement positions with the same surrounding environment, and retaining the remaining initial measurement positions as the measurement positions.

14. The method for establishing the etching deviation compensation model as described in claim 13, characterized in that, Filtering multiple initial measurement positions includes: grouping the spatial points corresponding to the initial measurement positions in multiple sampled images into the same coordinate system space; Obtain the spacing between spatial points in the coordinate system space; When the spacing is less than or equal to the spacing threshold, remove either of the two corresponding spatial points and retain the remaining spatial points; Otherwise, the corresponding spatial point is retained.

15. The method for establishing the etching deviation compensation model as described in claim 1, characterized in that, Based on the graphic contour, the etching deviation value at the measurement position is obtained, including: obtaining the simulated exposure pattern corresponding to the measurement pattern; The size difference between the graphic contour and the simulated exposure graphic at the measurement location is used as the etching deviation value.

16. The method for establishing the etching deviation compensation model as described in claim 13, characterized in that, Based on the etching deviation value at the measurement location, an etching deviation compensation model is established, including: obtaining the graphic feature value of the measurement location; The etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement location.

17. The method for establishing the etching deviation compensation model as described in claim 16, characterized in that, The etching deviation compensation model is obtained by combining the etching deviation compensation value and the graphic feature value at the measurement position, including: performing function fitting on the etching deviation compensation value and the graphic feature value at the measurement position to obtain the coefficients corresponding to the graphic feature value, and the coefficients and the graphic feature value constitute the etching deviation compensation model.

18. A system for establishing an etching deviation compensation model, characterized in that, include: A measurement structure providing module is used to provide a measurement structure, including multiple measurement patterns, wherein the measurement patterns are etched patterns; The sampling location selection module is used to select the sampling location of the measurement structure, wherein a portion of the measurement pattern at the sampling location represents the feature information of the measurement structure; Selecting a portion of the initial sampling locations as sampling locations, the sum of the graphic feature information of the portion of the initial sampling locations covers the feature information of the measured structure; Among the initial sampling locations selected as sampling locations, based on the distribution of feature information in the initial sampling images, the initial sampling locations corresponding to multiple initial sampling images whose feature information distribution covers the feature information of the measured structure are selected as sampling locations. A sampling image acquisition module is used to acquire multiple sampling images at the sampling location; The graphic contour acquisition module is used to obtain the graphic contour of the measurement graphic corresponding to the sampling position based on multiple sampled images; A measurement position acquisition module is used to acquire multiple measurement positions on the graphic contour; The etching deviation value acquisition module is used to acquire the etching deviation value at the measurement position based on the graphic contour. The etching deviation compensation model establishment module is used to establish the etching deviation compensation model based on the etching deviation value at the measurement position.

19. An etching deviation compensation model established using the method described in any one of claims 1-17.

20. An etching deviation compensation method based on an etching deviation compensation model established by the method as described in any one of claims 1-17.

21. A device, characterized in that, It includes at least one memory and at least one processor, the memory storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method for establishing an etching deviation compensation model as described in any one of claims 1-17.

22. A storage medium, characterized in that, The storage medium stores one or more computer instructions, which are used to implement the method for establishing the etching deviation compensation model as described in any one of claims 1-17.

Images