Quality evaluation methods for different OPC configuration models

Abstract

The present invention discloses a quality evaluation method for different OPC configuration models, comprising: step 1, providing different OPC tool software. Step 2, intercepting the structure area to be evaluated in the layout. Step 3, using each OPC tool software in turn to simulate each structure graphic to be evaluated in the structure area to be evaluated and obtain corresponding simulation graphics. Step 4, superimposing each simulation graphics together. Step 5, performing grid division, setting the weight of each grid point and setting the grid point information corresponding to each simulation graphic; performing matrix-vector processing on the grid point information corresponding to each simulation graphic and obtaining a matrix vector. Step 6, establishing a similarity evaluation function, substituting each matrix vector of the simulation graphics corresponding to the two OPC tool software to be compared into the similarity evaluation function to obtain a similarity evaluation value, and judging the accuracy of the model of the OPC tool software corresponding to the structure area to be evaluated based on the similarity evaluation value.

Description

Technical Field

[0001] The present invention relates to a semiconductor integrated circuit manufacturing process, and in particular to a quality evaluation method for different configuration models of optical proximity correction (OPC). Background Art

[0002] To evaluate the stability of different OPC software simulations for different structures, structural graphics of the area to be verified are often captured and simulated to assess the accuracy of the structural model. Existing methods involve manually capturing the area to be verified, selecting the corresponding process information, generating configuration files for OPC simulations using different software, and then manually overlaying the generated OPC simulation results onto the same interface for qualitative accuracy evaluation. This approach is extremely time-consuming, labor-intensive, and prone to errors. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a quality evaluation method for different OPC configuration models, which can automatically determine the accuracy of model simulation of different OPC tool software in different structural areas, and can further realize the automated evaluation and analysis of OPC models of OPC tool software suitable for different products, different process platforms and different manufacturers, and improve the versatility.

[0004] To solve the above technical problems, the present invention provides a quality evaluation method for different OPC configuration models, comprising the following steps:

[0005] Step 1: Provide different OPC tool software, each of which has a different OPC model.

[0006] Step 2: intercepting a structure region to be evaluated in the layout, wherein the structure region to be evaluated has more than one structure graphics to be evaluated.

[0007] Step 3: using each of the OPC tool software in turn to simulate each of the structure graphics to be evaluated in the structure area to be evaluated and obtain corresponding simulation graphics.

[0008] Step 4: Overlay the simulation graphics corresponding to the OPC tool software.

[0009] Step 5. Grid the simulation graphics corresponding to each of the OPC tool software, set the weight of each grid point, and set the grid point information corresponding to the simulation graphics corresponding to each of the OPC tool software; perform matrix-vector processing on the grid point information corresponding to the simulation graphics corresponding to each of the OPC tool software and obtain the matrix vectors corresponding to the simulation graphics corresponding to each of the OPC tool software.

[0010] Step 6: Establish a similarity evaluation function, substitute the matrix vectors of the simulation graphics corresponding to the two OPC tool software to be compared into the similarity evaluation function to obtain a similarity evaluation value, and judge the accuracy of the model of the OPC tool software corresponding to the structural area to be evaluated based on the similarity evaluation value.

[0011] A further improvement is that, in step 2, the layout includes graphics of multiple process levels, and the structural graphics to be evaluated are located in the selected process level.

[0012] A further improvement is that, in step 2, the intercepted area is larger than the area of ​​the structure to be evaluated.

[0013] A further improvement is that the distance between the edge of each of the intercepted areas and the edge of the corresponding structure area to be evaluated is greater than 1 micron.

[0014] A further improvement is that in step three, before performing the simulation, it also includes extracting configuration information and process condition information of the product corresponding to the layout.

[0015] Each of the OPC tool software receives the configuration information and the process condition information and generates a configuration file corresponding to each of the OPC tool software, and then performs the simulation.

[0016] A further improvement is that before step 3, the following is also included:

[0017] A configuration template library is established, wherein the configuration template library includes configuration file templates corresponding to each of the OPC tool software.

[0018] In step three, the configuration information and the process condition information are automatically extracted; after the configuration information and the process condition information are extracted, the configuration information and the process condition information are automatically filled into the configuration file template to form the configuration file.

[0019] A further improvement is that the configuration template library also includes a simulation sandbox corresponding to each of the OPC tool software.

[0020] In step three, after the configuration file is formed, the OPC tool software automatically completes the simulation in the simulation sandbox.

[0021] A further improvement is that, in step 4, the simulation graphics corresponding to each of the OPC tool software are pasted on the current interface to achieve superposition.

[0022] A further improvement is that, in step five, each of the grid points formed by the grid division is a square with the same side length.

[0023] A further improvement is that the side length of each of the lattice points is less than or equal to 1 nm.

[0024] A further improvement is that the setting of the grid information corresponding to the simulation graphics corresponding to each of the OPC tool software includes:

[0025] The grid point information of the grid points completely covered by the simulation graphic is set to 0.

[0026] The grid point information of the grid points that are not completely covered by the simulation graphic is set to 0.

[0027] The grid point information of the grid points whose area covered by the simulation graphic is less than 50% is set to 1.

[0028] The grid point information of the grid points of which the area covered by the simulation graphic is greater than or equal to 50% is set to 2.

[0029] A further improvement is that, in step five, the weights are set according to the location area, and the weights corresponding to the grid points in the same weight area are the same.

[0030] A further improvement is that, in step 5, the matrix vectorization processing includes:

[0031] In the superimposed graphic of the simulation graphic, rectangles are used to divide the area of ​​the superimposed graphic into multiple simulation areas.

[0032] The grid point information of the simulation graphics corresponding to each OPC tool software in each simulation area is formed into matrix information according to the position of the grid point in the rectangle corresponding to the simulation area.

[0033] Each of the grid point information of each of the matrix information is converted into a one-dimensional vector.

[0034] The information of the simulation graphics corresponding to each of the OPC tool software is represented by a one-dimensional vector corresponding to each of the simulation areas, thereby realizing the matrix-vector processing.

[0035] A further improvement is that the similarity evaluation function is expressed as follows:

[0036]

[0037] Among them, Eva A-B Represents the similarity evaluation values ​​corresponding to the two OPC tool software;

[0038] A corresponds to the number of the OPC tool software;

[0039] B corresponds to the serial number of another OPC tool software;

[0040] ai j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered A;

[0041] bi j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered B;

[0042] wti j Represents ai j The corresponding weight of the grid point;

[0043] i represents the number of the simulation area corresponding to the one-dimensional vector;

[0044] j represents the number of the element value in the one-dimensional vector.

[0045] ni represents the maximum number of elements of the matrix information obtained in the simulation area corresponding to the one-dimensional vector corresponding to i.

[0046] A further improvement is that, in step six, when the number of the OPC tool software is greater than three, each of the OPC tool software is compared with each other and a corresponding similarity evaluation value is formed.

[0047] Further improvements include:

[0048] The similarity evaluation values ​​formed by comparing the OPC tool software in pairs are arithmetic averaged to obtain a similarity evaluation average value.

[0049] The similarity evaluation average value is compared with a similarity evaluation threshold.

[0050] If the similarity evaluation average value is less than the similarity evaluation threshold value, it indicates that the accuracy of the models of the OPC tool software corresponding to the structural area to be evaluated is questionable, and the structural area to be evaluated is an inaccurate area;

[0051] If the similarity evaluation average value is greater than or equal to the similarity evaluation threshold value, it means that the accuracy of the models of the OPC tool software corresponding to the structural area to be evaluated meets the requirements.

[0052] A further improvement is that the method further includes: replacing the structural area to be evaluated in step 2, and then repeating steps 3 to 6 to determine the accuracy of the model of the OPC tool software corresponding to the replaced structural area to be evaluated.

[0053] A further improvement is that the method further includes: selecting the inaccurate area, and using the graphic structure in the inaccurate area to guide the establishment or optimization of subsequent models of each of the OPC tool software.

[0054] The present invention superimposes simulation graphics formed by simulating the structural graphics of the structural area to be evaluated by different OPC tool software, then adopts grid division and sets grid point weights, and sets grid point information for the grid points formed by grid division according to the simulation graphics, performs matrix-vector processing on the grid point information and obtains each matrix vector. In this way, by comparing the matrix vectors of the simulation graphics of the OPC tool software in pairs and substituting them into the similarity evaluation function, a similarity evaluation value can be obtained. Based on the similarity evaluation value, the quality of the corresponding OPC tool software can be judged. Here, the quality of the OPC tool software mainly refers to the accuracy of the model of the OPC tool software corresponding to the structural area to be evaluated. Since the information of the matrix vectors is obtained by calculation based on the simulation graphics, the similarity evaluation value can also be automatically obtained by substituting the information of the matrix vectors into the similarity evaluation function. Therefore, the quality of the corresponding OPC tool software can be automatically judged based on the calculation, that is, the accuracy of the model simulation of different OPC tool software corresponding to the structural area to be evaluated can be automatically judged. The higher the simulation accuracy, the better the quality of the corresponding OPC tool software.

[0055] The present invention can also use different OPC tool software to automatically simulate the structural areas to be evaluated in the layout of different products and different process platforms. For example, it can automatically intercept the structural areas to be evaluated at the corresponding process level on the layout, automatically extract the configuration information and process condition information of the product, and establish configuration file templates and simulation sandboxes corresponding to each OPC tool software. The automatically extracted configuration information and process condition information can be automatically filled into the configuration file template to form a configuration file and automatically simulated through the simulation sandbox. Therefore, the present invention can realize the automation of the entire process and can be applicable to OPC tool software of different products, different process platforms and different manufacturers. Therefore, the present invention can further realize the automated evaluation and analysis of the OPC models of OPC tool software applicable to different products, different process platforms and different manufacturers and improve the versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0057] Figure 1 is a flow chart of a quality evaluation method for different OPC configuration models according to an embodiment of the present invention;

[0058] Figure 2 This is a flow chart of a quality evaluation method for different OPC configuration models according to a preferred embodiment of the present invention;

[0059] Figure 3 It is a graphic structure of a cutout area of ​​a quality evaluation method layout of different OPC configuration models according to an embodiment of the present invention;

[0060] Figure 4 Flowchart of the quality evaluation method for different OPC configuration models according to the preferred embodiment of the present invention for performing automated OPC simulation;

[0061] Figure 5 It is an overlay of multiple simulation graphics obtained by the quality evaluation method of different OPC configuration models according to an embodiment of the present invention;

[0062] Figure 6 The graph is a graph obtained by meshing a simulation graph and setting grid point information according to the quality evaluation method for different OPC configuration models in an embodiment of the present invention;

[0063] Figure 7 Yes Figure 6 The grid information of a simulation graphic is the matrix information formed by dividing the simulation area. DETAILED DESCRIPTION

[0064] like Figure 1 As shown in FIG, it is a flow chart of a quality evaluation method of different OPC configuration models according to an embodiment of the present invention; Figure 2 As shown in FIG, it is a flow chart of a quality evaluation method of different OPC configuration models according to a preferred embodiment of the present invention; Figure 4 FIG. 2 is a flow chart of a method for evaluating the quality of different OPC configuration models according to a preferred embodiment of the present invention for performing automated OPC simulation. The method for evaluating the quality of different OPC configuration models according to the embodiment of the present invention includes the following steps:

[0065] Step 1: Provide different OPC tool software, each of which has a different OPC model.

[0066] OPC tool software is usually provided by different vendors. Figure 2 The OPC tool software corresponding to three vendors is shown, namely vendor-A tool, vendor-B tool, and vendor-C tool.

[0067] In a preferred embodiment of the present invention, before the subsequent step three, the method further includes:

[0068] A configuration template library 101 is established, wherein the configuration template library 101 includes configuration file templates 102 corresponding to each of the OPC tool software. Figure 2 The three configuration file templates 102 corresponding to vendor-A tool, vendor-B tool, and vendor-C tool are displayed.

[0069] The configuration template library 101 also includes simulation sandboxes corresponding to the respective OPC tool software.

[0070] Step 2: Clipping a structure region to be evaluated in a mask, wherein the structure region to be evaluated has one or more structure patterns 105 to be evaluated.

[0071] like Figure 3 1 is a graphical structure of a cutout area of ​​a quality evaluation method layout for different OPC configuration models according to an embodiment of the present invention; Figure 3 In the figure, the structure area to be evaluated is located in the inner area of ​​the rectangular line 103. Figure 3 In FIG, a total of five structure graphics 105 to be evaluated are displayed, and the five structure graphics 105 to be evaluated are marked with 1, 2, 3, 4 and 5 respectively.

[0072] In the embodiment of the present invention, the layout includes graphics of multiple process levels, and the structure graphic to be evaluated 105 is located in a selected process level.

[0073] The intercepted area is larger than the structure area to be evaluated. Figure 3 In FIG, the intercepted area is the area inside the rectangular dotted line 104, and the area inside the rectangular dotted line 104 is also called the P-box or the patch part.

[0074] The area between the rectangular line 103 and the rectangular dashed line 104 is a boundary box, ie, a B-box.

[0075] The structure area to be evaluated inside the rectangular line 103 is also called an evaluation box, or E-box.

[0076] The distance between the edge of each cut area and the edge of the corresponding structure area to be evaluated is more than 1 micron, that is, Figure 3 The width d of the B-box in the middle is more than 1 micron. This is to avoid simulation failure in the boundary area. Therefore, it is necessary to cut off about 1 micron on one side at the verification structure.

[0077] Figure 2 In the example, step 2 corresponds to step S101, clipping the area to be simulated. The area is the clipped area clipped from the mask.

[0078] Figure 4 In the step 2, step 2 corresponds to step S101a, clip Mask, that is, clipping the clipped area from the Mask.

[0079] Step 3: Use the OPC tool software in sequence to simulate each of the to-be-evaluated structural graphics 105 in the to-be-evaluated structural area and obtain a corresponding simulation graphic.

[0080] In an embodiment of the present invention, before performing the simulation, the process further includes extracting configuration information and process condition information of the product corresponding to the layout.

[0081] In a preferred embodiment of the present invention, the configuration information and the process condition information are automatically extracted. Figure 2 In the step of automatically extracting the configuration information and the process condition information, the step corresponds to step S102, where information such as configuration and technique is extracted from the current product path. The configuration information and the process condition information are both product-related, and thus can be extracted from the current product path. Figure 4 In the embodiment, this step corresponds to step S102a, information such as configuration, technique.

[0082] In the embodiment of the present invention, each of the OPC tool software receives the configuration information and the process condition information and generates a configuration file corresponding to each of the OPC tool software, and then performs the simulation.

[0083] In a preferred embodiment of the present invention, after the configuration information and the process condition information are extracted, the configuration information and the process condition information are automatically filled into the configuration file template 102 to form the configuration file. Figure 2 In the example, the configuration information and the process condition information are automatically filled into the configuration file template 102, corresponding to step S103, Select the corresponding template and fill it. After filling, the configuration file is generated, i.e. Figure 2 Step S104, Generate configuration file.

[0084] Figure 4 In the process, the configuration information and the process condition information are automatically filled into the configuration file template 102, corresponding to step S103a, Fill. Figure 4It can be seen that the configuration information and the process condition information are shared by a plurality of the OPC tool software, and it is only necessary to automatically fill the configuration information and the process condition information into the configuration file template 102 of the corresponding OPC tool software. Figure 4 In the example, the configuration files generated are displayed in an expanded manner according to the different OPC tool software, namely S104a, Generate configuration file of A-tool; S104b, Generate configuration file of B-tool; S104c, Generate configuration file of C-tool.

[0085] In step three, after the configuration file is formed, the OPC tool software automatically completes the simulation in the simulation sandbox.

[0086] In a preferred embodiment of the present invention, Figure 2 As shown, the simulation step corresponds to step S105, Run OPC simulation. Combining the configuration file and the structure area to be evaluated, each structure graphic 105 to be evaluated in the structure area to be evaluated can be simulated and a simulation image corresponding to each structure graphic 105 to be evaluated can be obtained.

[0087] Step 4: Overlay the simulation graphics corresponding to the OPC tool software.

[0088] In the embodiment of the present invention, the simulation graphics corresponding to each of the OPC tool software are pasted on the current window to achieve superimposition.

[0089] Step 4 corresponds to Figure 2 In step S106, the simulation image superimposes the current window.

[0090] like Figure 5 , which is an overlay of multiple simulation graphics obtained by the quality evaluation method of different OPC configuration models according to an embodiment of the present invention; Figure 5 Shown in Figure 3The three simulation graphics corresponding to the two structural graphics to be evaluated 105 numbered 1 and 2 are represented by marks 106a, 106b and 106c respectively. Mark 106a represents the simulation graphic obtained by simulation using Vendor-A tool, mark 106b represents the simulation graphic obtained by simulation using Vendor-B tool, and mark 106c represents the simulation graphic obtained by simulation using Vendor-C tool.

[0091] Step 5: Meshing the simulation graphics corresponding to each of the OPC tool software.

[0092] like Figure 6 1 is a graph after meshing and setting grid point information of a simulation graph according to a quality evaluation method for different OPC configuration models according to an embodiment of the present invention. Figure 6 In the Figure 3 The structure graph to be evaluated 105 and the corresponding three simulation graphs 106a, 106b and 106c are shown in FIG.

[0093] In an embodiment of the present invention, each of the grid points 107 formed by the grid division is a square with the same side length. In some embodiments, the side length of each of the grid points 107 is less than or equal to 1 nm.

[0094] The weight of each grid point 107 is set, and the grid point information corresponding to the simulation graph corresponding to each OPC tool software is set.

[0095] Figure 6 The values ​​of the grid point information in the partial simulation area corresponding to A1 of the simulation graphic 106a are displayed in FIG. The settings of the grid point information corresponding to the simulation graphic corresponding to each OPC tool software include:

[0096] The grid point information of the grid points 107 completely covered by the simulation graphic is set to 0.

[0097] The grid point information of the grid points 107 that are not completely covered by the simulation pattern is set to 0.

[0098] The grid point information of the grid points 107 of which the area covered by the simulation graphic is less than 50% is set to 1.

[0099] The grid point information of the grid points 107 of which the area covered by the simulation graphic is greater than or equal to 50% is set to 2.

[0100] In the embodiment of the present invention, the weight is set according to the location area, and the weight corresponding to each grid point 107 in the same weight area is the same.

[0101] Matrix-vector processing is performed on the grid point information corresponding to the simulation graphics corresponding to each of the OPC tool software to obtain matrix vectors corresponding to the simulation graphics corresponding to each of the OPC tool software.

[0102] In an embodiment of the present invention, the matrix vectorization processing includes:

[0103] In the superimposed graphics of the simulation graphics, rectangles are used to divide the area of ​​the superimposed graphics into multiple simulation areas. Figure 6 A1 and A2 in the figure are the two simulation areas corresponding to the simulation graph 106a. Similarly, the simulation area corresponding to the simulation graph 106b at the same position as A1 can be represented by B1; the simulation area corresponding to the simulation graph 106b at the same position as A2 can be represented by B2, and so on. Using i to represent the rectangular number of the simulation area, each simulation area of ​​the simulation graph 106a can be represented by Ai, and the matrix obtained subsequently is also represented by Ai, and the one-dimensional vector formed by the expansion of the matrix Ai is represented by ai; each simulation area of ​​the simulation graph 106a can be represented by Bi, and the matrix obtained subsequently is also represented by Bi, and the one-dimensional vector formed by the expansion of the matrix Bi is represented by bi; each simulation area of ​​the simulation graph 106c can be represented by Ci, and the corresponding matrix is ​​also represented by Ci, and the one-dimensional vector formed by the expansion of the matrix Ci is represented by ci.

[0104] The grid point information of the simulation graphics corresponding to each OPC tool software in each simulation area is formed into matrix information according to the position of the grid point 107 in the rectangle corresponding to the simulation area. Figure 7 As shown, it is Figure 6 Matrix information formed by dividing the grid point information of a simulation graphic 106a according to the simulation area; Figure 7 The values ​​of each element of matrix A1 are given by Figure 6 The values ​​of the grid information in the simulation area corresponding to A1 are composed of: Figure 6 The values ​​of the grid point information in the simulation area corresponding to A2 are composed of the above; and so on, more matrices can be obtained according to actual conditions, such as matrices A3 and A4. Figure 7 In, M A1 Indicates the maximum number of rows in matrix A1, N A1 represents the maximum number of columns of matrix A1; similarly, M A2 Indicates the maximum number of rows in matrix A2, N A2 Indicates the maximum number of columns in matrix A2.

[0105] To facilitate subsequent processing, each grid point of each matrix information is converted into a one-dimensional vector. For example, each element in the matrix A1 is expanded into a one-dimensional vector, that is, each element in the matrix A1 is arranged into a one-dimensional structure, which is convenient for subsequent substitution into the similarity evaluation function for calculation.

[0106] The information of the simulation graphics corresponding to each of the OPC tool software is represented by a one-dimensional vector corresponding to each of the simulation areas, thereby realizing the matrix-vector processing.

[0107] Step 6: Establish a similarity evaluation function, substitute the matrix vectors of the simulation graphics corresponding to the two OPC tool software to be compared into the similarity evaluation function to obtain a similarity evaluation value, and judge the accuracy of the model of the OPC tool software corresponding to the structural area to be evaluated based on the similarity evaluation value.

[0108] In the embodiment of the present invention, the quality of the OPC tool software primarily refers to the accuracy of the model simulation of the OPC tool software. The model of the OPC tool software is also related to the structure of the structural area to be evaluated. For the same structural graph 105 to be evaluated, if the differences in the simulated graphs generated by different OPC tool software are small, it can be shown that the OPC tool software model corresponding to the structural area to be evaluated is accurate.

[0109] The similarity evaluation function is expressed as follows:

[0110]

[0111] Among them, Eva A-B Represents the similarity evaluation values ​​corresponding to the two OPC tool software;

[0112] A corresponds to the number of the OPC tool software;

[0113] B corresponds to the serial number of another OPC tool software;

[0114] ai j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered A;

[0115] bi j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered B;

[0116] wti j Represents ai j The corresponding weight of the grid point.

[0117] i represents the number of the simulation area corresponding to the one-dimensional vector, for example: Figure 7 The i corresponding to matrix A1 is 1, the i corresponding to matrix A2 is 2, the i corresponding to matrix A3 is 3, and the i corresponding to matrix A4 is 4; and matrices A1 and A2 are respectively Figure 6 The simulation areas A1 and A2 in FIG. 106b correspond to each other. Similarly, the matrix B1 corresponding to the simulation diagram 106b is Figure 6 The area in the matrix A1 is the same as that in the matrix A1, and the corresponding i is also 1; the matrix B2 corresponding to the simulation diagram 106b is Figure 6 The area in is the same as that of matrix A2, and the corresponding i is also 2. Similarly, more i can be obtained. In the formula of the similarity evaluation function described above, only the cases where i is 1 and 2 are expanded, and other values ​​of i are omitted.

[0118] j represents the number of the element value in the one-dimensional vector.

[0119] ni represents the maximum number of elements of the matrix information obtained from the simulation area corresponding to the one-dimensional vector corresponding to i. In the formula of the similarity evaluation function above, n1 represents Figure 7 The product of the rows and columns of the matrix A1 is M A1 ×N A1 ; n2 means Figure 7 The sum of the rows and columns of the matrix A2 is M A2 ×N A2 The rows and columns of matrix B1 are the same as those of matrix A1, so in fact n1 is also the product of the rows and columns of matrix B1, n2 is also the product of the rows and columns of matrix B2, and so on.

[0120] Eva A-B The corresponding similarity evaluation value can represent the similarity between Vendor-A tool and Vendor-B tool, and the range of the similarity evaluation value is between 0 and 1.

[0121] In a preferred embodiment of the present invention, when there are more than three OPC tool software, each of the OPC tool software is compared with each other and the corresponding similarity evaluation value is formed. For example: in addition to comparing the similarity of Vendor-A tool and Vendor-B tool to form Eva A-B In addition to the corresponding similarity evaluation value, it also includes: comparing the similarity of Vendor-A tool and Vendor-C tool to form Eva A-c The corresponding similarity evaluation value, and the similarity comparison between Vendor-B tool and Vendor-C tool to form Eva B-C The corresponding similarity evaluation value.

[0122] The similarity evaluation values ​​formed by the pairwise comparison of the OPC tool software are arithmetic averaged to obtain the similarity evaluation average value. The calculation formula of the similarity evaluation average value can be expressed as:

[0123]

[0124] represents the average value of the similarity evaluation.

[0125] The similarity evaluation average value is compared with a similarity evaluation threshold value. The similarity evaluation threshold value can be represented by Ex.

[0126] If the similarity evaluation average value is less than the similarity evaluation threshold value, it means that the accuracy of the model of each OPC tool software corresponding to the structure area to be evaluated is questionable, and the structure area to be evaluated is an inaccurate area. When it is less than Ex, it means that the simulation differences between different tools are large, and the structural graphics will be returned to the library, indicating that the model accuracy of this structural graphics is questionable; in addition, the different structural shapes returned also provide a reference for subsequent optimization of the OPC model.

[0127] If the similarity evaluation average value is greater than or equal to the similarity evaluation threshold value, it means that the accuracy of the models of the OPC tool software corresponding to the structural area to be evaluated meets the requirements.

[0128] In an embodiment of the present invention, the structural area to be evaluated can be further changed to obtain the accuracy of the model of each OPC tool software corresponding to other structural areas to be evaluated, including:

[0129] In step 2, the structural area to be evaluated is replaced, and then steps 3 to 6 are repeated to determine the accuracy of the model of the OPC tool software corresponding to the replaced structural area to be evaluated.

[0130] In an embodiment of the present invention, the method further includes: selecting the inaccurate area, and using the graphic structure in the inaccurate area to guide the establishment or optimization of subsequent models of each of the OPC tool software.

[0131] In an embodiment of the present invention, simulation graphics formed by simulating the structural graphics 105 of the structural area to be evaluated by different OPC tool software are superimposed, and then grid division is performed and grid point 107 weights are set, and grid point information is set for the grid points 107 formed by the grid division according to the simulation graphics, and the grid point information is matrix-vectorized to obtain matrix vectors. In this way, similarity evaluation values ​​can be obtained by comparing the matrix vectors of the simulation graphics of the OPC tool software with each other and substituting them into a similarity evaluation function. Based on the similarity evaluation value, the quality of the corresponding OPC tool software can be judged. Here, the quality of the OPC tool software mainly refers to the accuracy of the model of the OPC tool software corresponding to the structural area to be evaluated. Since the information of the matrix vectors is calculated based on the simulation graphics, the similarity evaluation value can also be automatically obtained by substituting the information of the matrix vectors into the similarity evaluation function. Therefore, the quality of the corresponding OPC tool software can be automatically judged based on the calculation, that is, the accuracy of the model simulation of different OPC tool software corresponding to the structural area to be evaluated can be automatically judged. The higher the simulation accuracy, the better the quality of the corresponding OPC tool software.

[0132] The embodiment of the present invention can also use different OPC tool software to automatically simulate the structural areas to be evaluated in the layout of different products and different process platforms. For example, it can automatically intercept the structural areas to be evaluated at the corresponding process level on the layout, automatically extract the configuration information and process condition information of the product, and establish a configuration file template 102 and a simulation sandbox corresponding to each OPC tool software. The automatically extracted configuration information and process condition information can be automatically filled into the configuration file template 102 to form a configuration file and automatically simulated through the simulation sandbox. Therefore, the embodiment of the present invention can realize the automation of the entire process and can be applicable to OPC tool software of different products, different process platforms and different manufacturers. Therefore, the embodiment of the present invention can further realize the automated evaluation and analysis of the OPC model of the OPC tool software applicable to different products, different process platforms and different manufacturers and improve the versatility.

[0133] As can be seen from the above, the embodiment of the present invention can realize the automatic configuration and simulation of OPC software parameter information between different vendors, and evaluate the accuracy of models under different structures based on the similarity (Cosine Similarity) method. That is, first establish the OPC standardized configuration file and simulation sandbox corresponding to the vendor, and automatically fill in the template and simulate according to the configuration information and process conditions provided by the user; then generate the results to the user folder, and automatically overlay the local simulation diagram on the current interface; finally, perform matrix-vector processing on the simulation graphics and set the weight coefficient, and quantitatively study the evaluation function to judge the accuracy and stability of the model. The method of the embodiment of the present invention saves time and effort, and more accurately and conveniently realizes local position simulation and model stability analysis between different vendors.

[0134] The present invention has been described in detail above by means of specific embodiments, but these do not constitute limitations of the present invention. Without departing from the principles of the present invention, those skilled in the art may make many variations and improvements, which should also be considered as the scope of protection of the present invention.

Claims

1. A quality evaluation method for different OPC configuration models, characterized in that: The steps include: Step 1: providing different OPC tool software, each of the OPC tool software having a different OPC model; Step 2: intercepting a structure region to be evaluated in the layout, wherein the structure region to be evaluated has more than one structure pattern to be evaluated; Step 3: using each of the OPC tool softwares in turn to simulate each of the structure graphics to be evaluated in the structure area to be evaluated and obtain corresponding simulation graphics; Step 4: Overlaying the simulation graphics corresponding to the OPC tool software; Step 5: Meshing the simulation graphics corresponding to each of the OPC tool software, setting the weight of each grid point, and setting grid point information corresponding to the simulation graphics corresponding to each of the OPC tool software; Performing matrix-vector processing on the grid point information corresponding to the simulation graphics corresponding to each of the OPC tool softwares to obtain matrix vectors corresponding to the simulation graphics corresponding to each of the OPC tool softwares; Step 6: Establish a similarity evaluation function, substitute the matrix vectors of the simulation graphics corresponding to the two OPC tool software to be compared into the similarity evaluation function to obtain a similarity evaluation value, and judge the accuracy of the model of the OPC tool software corresponding to the structural area to be evaluated based on the similarity evaluation value.

2. The quality evaluation method of different OPC configuration models according to claim 1, characterized in that: In step 2, the layout includes graphics of multiple process levels, and the structural graphics to be evaluated are located in the selected process level.

3. The quality evaluation method of different OPC configuration models according to claim 2, characterized in that: In step 2, the intercepted area is larger than the structure area to be evaluated.

4. The quality evaluation method of different OPC configuration models according to claim 3, characterized in that: The distance between the edge of each of the cutout regions and the edge of the corresponding structure region to be evaluated is greater than 1 micron.

5. The quality evaluation method of different OPC configuration models according to claim 2, characterized in that: In step three, before performing the simulation, the configuration information and process condition information of the product corresponding to the layout are extracted; Each of the OPC tool software receives the configuration information and the process condition information and generates a configuration file corresponding to each of the OPC tool software, and then performs the simulation.

6. The quality evaluation method of different OPC configuration models according to claim 5, characterized in that: Before step 3, it also includes: Establishing a configuration template library, wherein the configuration template library includes configuration file templates corresponding to each of the OPC tool software; In step three, the configuration information and the process condition information are automatically extracted; after the configuration information and the process condition information are extracted, the configuration information and the process condition information are automatically filled into the configuration file template to form the configuration file.

7. The quality evaluation method of different OPC configuration models according to claim 6, characterized in that: The configuration template library also includes a simulation sandbox corresponding to each of the OPC tool software; In step three, after the configuration file is formed, the OPC tool software automatically completes the simulation in the simulation sandbox.

8. The quality evaluation method for different OPC configuration models according to claim 1, wherein: In step 4, the simulation graphics corresponding to each of the OPC tool software are pasted on the current interface to achieve superposition.

9. The quality evaluation method of different OPC configuration models according to claim 1, characterized in that: In step five, each of the grid points formed by the grid division is a square with the same side length.

10. The quality evaluation method of different OPC configuration models according to claim 9, characterized in that: The side length of each of the lattice points is less than or equal to 1 nm.

11. The quality evaluation method of different OPC configuration models according to claim 9, characterized in that: The setting of the grid information corresponding to the simulation graphics corresponding to each of the OPC tool software includes: Setting the grid point information of the grid points completely covered by the simulation graphic to 0; Setting the grid point information of the grid points that are not completely covered by the simulation graphic to 0; Setting the grid point information of the grid points whose area covered by the simulation graphic is less than 50% to 1; The grid point information of the grid points of which the area covered by the simulation graphic is greater than or equal to 50% is set to 2.

12. The quality evaluation method for different OPC configuration models according to claim 11, characterized in that: In step five, the weight is set according to the location area, and the weight corresponding to each grid point in the same weight area is the same.

13. The quality evaluation method for different OPC configuration models according to claim 12, characterized in that: In step 5, the matrix vectorization process includes: In the superimposed graphic of the simulation graphic, a region of the superimposed graphic is divided into a plurality of simulation regions using rectangles; Forming matrix information of the grid point information of the simulation graphics corresponding to the OPC tool software in each simulation area according to the positions of the grid points in the rectangle corresponding to the simulation area; Converting each of the grid points of the matrix information into a one-dimensional vector; The information of the simulation graphics corresponding to each of the OPC tool software is represented by a one-dimensional vector corresponding to each of the simulation areas, thereby realizing the matrix-vector processing.

14. The quality evaluation method of different OPC configuration models according to claim 13, characterized in that: The similarity evaluation function is expressed as follows: Among them, Eva A-B Represents the similarity evaluation values ​​corresponding to the two OPC tool software; A corresponds to the number of the OPC tool software; B corresponds to the serial number of another OPC tool software; ai j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered A; bi j Represents the element value in the one-dimensional vector corresponding to the OPC tool software numbered B; wti j Represents ai j The corresponding weight of the grid point; i represents the number of the simulation area corresponding to the one-dimensional vector; j represents the number of the element value in the one-dimensional vector; ni represents the maximum number of elements of the matrix information obtained in the simulation area corresponding to the one-dimensional vector corresponding to i.

15. The quality evaluation method of different OPC configuration models according to any one of claims 1 to 14, characterized in that: In step 6, when the number of the OPC tool software is greater than 3, each of the OPC tool software is compared with each other and a corresponding similarity evaluation value is generated.

16. The quality evaluation method of different OPC configuration models according to claim 15, characterized in that: Also includes: Performing arithmetic averaging on the similarity evaluation values ​​formed by pairwise comparison of the OPC tool software to obtain a similarity evaluation average value; comparing the similarity evaluation average value with a similarity evaluation threshold; If the similarity evaluation average value is less than the similarity evaluation threshold value, it indicates that the accuracy of the models of the OPC tool software corresponding to the structural area to be evaluated is questionable, and the structural area to be evaluated is an inaccurate area; If the similarity evaluation average value is greater than or equal to the similarity evaluation threshold value, it means that the accuracy of the models of the OPC tool software corresponding to the structural area to be evaluated meets the requirements.

17. The quality evaluation method of different OPC configuration models according to claim 16, characterized in that: Also includes: In step 2, the structural area to be evaluated is replaced, and then steps 3 to 6 are repeated to determine the accuracy of the model of the OPC tool software corresponding to the replaced structural area to be evaluated.

18. The quality evaluation method of different OPC configuration models according to claim 17, characterized in that: Also includes: The inaccurate area is selected, and the graphic structure in the inaccurate area is used to guide the establishment or optimization of the subsequent models of the OPC tool software.

Images