Optical proximity correction (OPC) segmentation method based on photoetching model
Through the OPC slicing method based on the lithography model, the problems of low OPC correction accuracy and long adjustment time in the prior art are solved, high-precision OPC correction and saving debugging time are achieved, and verification error rate is reduced.
Patent Information
- Application Number
- CN202510105738.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-22
- Publication Date
- 2025-03-25
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
When facing complex layouts and high-density process nodes, the existing OPC slicing method is difficult to improve the correction accuracy, and requires a lot of time to debug the slicing rules, resulting in a high verification error rate.
Using the OPC slicing method based on the lithography model, by reading the optimized edge position information in the initial layout, using the lithography model to simulate and generate contour information, performing the first correction and introducing a cost function related to the light intensity, the sampling points are dynamically adjusted to optimize the slicing until the convergence requirements are met.
It improves OPC correction accuracy, saves time to adjust segmentation rules, reduces OPC verification error rate, and achieves the effect of fine segmentation in key areas and meets manufacturing requirements.
Smart Images
Figure CN119689775A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a semiconductor integrated circuit manufacturing method, in particular to an OPC cutting method based on a photolithography model. Background Art
[0002] At present, as a kind of resolution enhancement technique (RET), optical proximity correction (OPC) technology is widely used in the manufacturing process of integrated circuit chips, and it becomes indispensable as the feature line width continues to decrease. In the mainstream OPC process, the slicing process is based on recipes, or rules. As the process nodes continue to shrink, the complexity and density of the layout continue to increase, forcing engineers to spend a lot of time debugging recipes to cover various complex two-dimensional graphics in order to eliminate various manufacturing defects caused by improper slicing. Faced with this challenge, research on slicing based on models, namely lithography models, has gradually begun to receive attention. Summary of the invention
[0003] The technical problem to be solved by the present invention is to provide an OPC segmentation method based on a lithography model, which can improve the OPC correction accuracy and save the segmentation rule adjustment time, and can effectively reduce the error rate of OPC verification.
[0004] To solve the above technical problems, the OPC cutting method based on the lithography model provided by the present invention comprises the following steps:
[0005] Step 1: Read the position information of the optimized edge of the graphics to be segmented and optimized in the initial layout, and use the lithography model to simulate the graphics to be segmented and optimized to generate the corresponding contour information of the optimized edge.
[0006] Step 2: Based on the contour information, a first correction is performed on the overall movement of each optimized edge of the graphics to be segmented and optimized, and a cost function related to light intensity is introduced.
[0007] The correction order in the first correction is: first correct the long side and then correct the short side, so that points with one-dimensional gradient of the cost function greater than 0 and one-dimensional gradient less than 0 can appear in the sampling points to ensure effective segmentation later.
[0008] Step 3: Select an initial sampling frequency to sample the light intensity value of the optimized edge, calculate the negative value of the gradient of the cost function at each sampling point, and use interpolation to obtain the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and troughs.
[0009] Step 4: based on the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and the troughs, the optimized edge is segmented and the sampling points are adjusted, so that the adjusted sampling points are closer to the peaks or the troughs than the sampling points before the adjustment.
[0010] Step 5: extract the adjusted light intensity value at each sampling point and calculate the corresponding offset of each optimized segment of the optimized edge after optimization.
[0011] Step 6: Repeat steps 3 to 5 until convergence requirements are met.
[0012] A further improvement is that in step 1, after the contour information is generated, the step further includes using OPC software to extract all spatial light intensity curves of the graphics to be segmented and optimized and observing the light intensity distribution of all sampling point positions on the spatial light intensity curves.
[0013] A further improvement is that in step 2, the formula of the cost function is:
[0014] Where C represents the cost function, r i represents the i-th sampling point, i represents the number of the sampling point, n represents the maximum value of i, I represents the light intensity value, I0 represents the average light intensity value of the optimized edge, I(r i ) represents r i The light intensity value at .
[0015] The gradient formula of the cost function along the normal direction of the i-th sampling point is: in, Represented as r i The normal direction corresponding to the line segment.
[0016] A further improvement is that in step 4, the segmentation includes cutting the optimized edge into a plurality of segmented sections and moving each of the segmented sections to achieve a second correction.
[0017] A further improvement is that the first correction is implemented using OPC software.
[0018] A further improvement is to use OPC software to implement the second correction.
[0019] A further improvement is that the initial layout includes multiple layers; in step 1, the lithography models used by the initial layouts at different layers are independently set according to the layers.
[0020] A further improvement is that in step 4, in the process of adjusting the sampling points, the sampling points are also adjusted according to the geometric environment.
[0021] A further improvement is that adjusting the sampling point according to the geometric environment includes:
[0022] The number of sampling points is reduced in non-critical layout areas to reduce the amount of calculation and save running time.
[0023] A further improvement is that adjusting the sampling point according to the geometric environment further includes:
[0024] In the critical layout area, it is ensured that the number of sampling points can meet the requirements of OPC correction quality.
[0025] A further improvement is that the key layout area includes: a corner area and a line end area.
[0026] A further improvement is that, in step one, the initial layout is read from the GDS layout file.
[0027] A further improvement is that, in step six, satisfying the convergence requirement includes: the offset corresponding to each segment of the optimized edge after optimization is less than or equal to a set value.
[0028] The OPC segmentation of the present invention is realized based on the lithography model. After the lithography model is used for simulation to obtain the corresponding contour information, the first correction of the overall movement of the optimized edge is first performed, and the correction order is to correct the long edge first and then the short edge. In this way, points with a one-dimensional gradient of the cost function greater than 0 and a one-dimensional gradient less than 0 can all appear in the sampling points, thereby ensuring that effective segmentation can be performed subsequently; then the segmentation, sampling point adjustment and the offset of the segmented segment after optimization are obtained according to the light intensity value extracted at the sampling point are repeated. In the repeated iterative steps, the segmentation method and the sampling point can be changed dynamically, for example, according to different lithography models and geometric environments. Finally, smaller line segments can be cut out in key areas and the line segments meet the minimum size requirements of layout manufacturing, thereby improving the quality of OPC correction; and long line segments are saved in places where fine segmentation is not required, thereby saving the running amount and running time. Finally, the present invention can improve the quality of OPC correction while saving the time for debugging segmentation rules, and can effectively reduce the error rate of OPC verification. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:
[0030] Figure 1 is a flow chart of an OPC slicing method based on a lithography model according to an embodiment of the present invention;
[0031] Figure 2 It is a contour information graph generated in step 1 of the OPC segmentation method based on the lithography model in an embodiment of the present invention;
[0032] Figure 3 The edge before and after the first correction of the OPC cutting method based on the lithography model in the embodiment of the present invention Figure 2 Light intensity distribution curve of the long side of center A;
[0033] Figure 4 is the layout segmentation result of four iterations of the OPC segmentation method based on the lithography model in the embodiment of the present invention;
[0034] Figure 5 It is the simulation result of the layout of 4 iterations of the OPC segmentation method based on the lithography model in the embodiment of the present invention;
[0035] Figure 6 is a comparison curve of the number of cuts of two layouts by the OPC cutting method based on the lithography model in the embodiment of the present invention and the existing method;
[0036] Figure 7 It is a comparative bar chart of the number of OPCV errors after segmentation of two layouts by the OPC segmentation method based on the lithography model in an embodiment of the present invention and the existing method. DETAILED DESCRIPTION
[0037] like Figure 1 FIG. 1 is a flow chart of an OPC segmentation method based on a lithography model according to an embodiment of the present invention. The OPC segmentation method based on a lithography model according to an embodiment of the present invention comprises the following steps:
[0038] Step 1: Read the position information of the optimized edge of the graphic to be segmented and optimized 102 in the initial layout, and use the lithography model to simulate the graphic to be segmented and optimized 102 to generate the corresponding contour information 103 of the optimized edge.
[0039] In the embodiment of the present invention, after the outline information 103 is generated, the process further includes extracting all spatial light intensity curves of the graphics 102 to be segmented and optimized using OPC software and observing the light intensity distribution of all sampling points on the spatial light intensity curves.
[0040] In an embodiment of the present invention, the initial layout includes multiple layers; the lithography models used by the initial layouts at different layers are independently set according to the layers.
[0041] The initial layout is read from a GDS layout file.
[0042] like Figure 2 As shown, it is a profile information 103 diagram generated in step 1 of the OPC segmentation method based on the lithography model according to an embodiment of the present invention; Figure 2 A partial area of an initial layout is shown and marked with a marker 101. Figure 2The initial layout 101 includes the graphics 102 to be segmented and optimized. The corresponding contour information 103 can be obtained by performing OPC simulation.
[0043] Step 1 corresponds to Figure 1 In step S1, the optimized edge position information of the graphics to be segmented and optimized in the layout is read, and a model simulation is performed to generate contour information.
[0044] Step 2: Based on the contour information 103, the optimized edges of the graphics 102 to be segmented and optimized are subjected to a first correction of overall movement and a cost function related to light intensity is introduced.
[0045] The correction order in the first correction is: first correct the long side and then correct the short side, so that points with one-dimensional gradient of the cost function greater than 0 and one-dimensional gradient less than 0 can appear in the sampling points to ensure effective segmentation later.
[0046] In the embodiment of the present invention, OPC software is used to implement the first calibration.
[0047] like Figure 3 FIG. 1 is an example of an OPC slicing method based on a photolithography model before and after the first correction of the embodiment of the present invention. Figure 2 The light intensity distribution curve of the long side of A in the middle; Curve 104 is the light intensity distribution curve of the long side of A before the first correction. Figure 2 The light intensity distribution curve of the long side of A in the middle, curve 105 is the light intensity distribution curve along the long side of A after the first correction. Figure 2 Light intensity distribution curve of the long side of center A.
[0048] In the embodiment of the present invention, the formula of the cost function is: C = ∑ i [(I(r i )-I0) 2 ], i=1 to n.
[0049] The formula of the cost function is:
[0050] Where C represents the cost function, r i represents the i-th sampling point, i represents the number of the sampling point, n represents the maximum value of i, I represents the light intensity value, I0 represents the average light intensity value of the optimized edge, I(r i ) represents r i The light intensity value at
[0051] The gradient formula of the cost function along the normal direction of the i-th sampling point is: in, Represented as r iThe normal direction corresponding to the line segment.
[0052] Step 2 corresponds to Figure 1 In step S2, based on the Contour information, the edges of the undivided polygons are first corrected and moved once, first correcting the long edges and then the short edges, so that and All points can appear in the sampling points.
[0053] Step 3: Select an initial sampling frequency to sample the light intensity value of the optimized edge, calculate the negative value of the gradient of the cost function at each sampling point, and use interpolation to obtain the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and troughs.
[0054] In some embodiments, the method further includes: calculating the inverse of the negative value of the gradient of the cost function at each sampling point, that is, Then use the interpolation method to get the point where the gradient of the optimized edge is equal to 0, that is point.
[0055] Step 3 corresponds to Figure 1 Step S3 in the above example selects a suitable initial sampling frequency and calculates the sampling point Value, obtained by interpolation The points and the locations of the peaks and troughs.
[0056] Step 4: based on the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and the troughs, the optimized edge is segmented and the sampling points are adjusted, so that the adjusted sampling points are closer to the peaks or the troughs than the sampling points before the adjustment.
[0057] In the embodiment of the present invention, the segmentation includes cutting the optimized edge into a plurality of segmented sections and moving each of the segmented sections to achieve a second correction.
[0058] The second calibration is implemented using OPC software.
[0059] In the embodiment of the present invention, the process of adjusting the sampling points further includes adjusting the sampling points according to a geometric environment.
[0060] Adjusting the sampling points according to the geometric environment includes:
[0061] The number of sampling points is reduced in non-critical layout areas to reduce the amount of calculation and save running time.
[0062] Adjusting the sampling points according to the geometric environment also includes:
[0063] In the critical layout area, it is ensured that the number of sampling points can meet the requirements of OPC correction quality.
[0064] The key layout areas include: corner areas and line end areas.
[0065] Step 5: extract the adjusted light intensity value at each sampling point and calculate the corresponding offset of each optimized segment of the optimized edge after optimization.
[0066] Steps 4 and 5 correspond to Figure 1 In step S4, the sampling points are divided and adjusted so that the sampling points are close to the peak or trough position. For non-critical layout areas, the sampling points are reduced to reduce the amount of calculation, save running time, and re-extract the light intensity value at the sampling point and calculate the corresponding offset after the optimization of the segment.
[0067] Step 6: Repeat steps 3 to 5 until convergence requirements are met.
[0068] In step six, satisfying the convergence requirement includes: the offset corresponding to each segment of the optimized edge after optimization is less than or equal to a set value.
[0069] Step 6 corresponds to Figure 1 In step S5, the above steps S3 to S4 are iterated repeatedly and the results are output, and each group of parameters is compared and the best parameter result is selected to output the final segmentation solution.
[0070] The OPC segmentation in the embodiment of the present invention is implemented based on the lithography model. After the lithography model is used for simulation to obtain the corresponding contour information 103, the first correction of the overall movement of the optimized edge is performed, and the correction order is to correct the long edge first and then the short edge. In this way, points with a one-dimensional gradient of the cost function greater than 0 and a one-dimensional gradient less than 0 can all appear in the sampling points, thereby ensuring that effective segmentation can be performed subsequently; then the segmentation, sampling point adjustment, and the offset of the segmented segment after optimization are obtained according to the light intensity value extracted at the sampling point are repeated. In the repeated iterative steps, the segmentation method and the sampling point can be changed dynamically, for example, according to different lithography models and geometric environments. Finally, smaller line segments can be cut out in key areas and the line segments meet the minimum size requirements of layout manufacturing, thereby improving the quality of OPC correction; and long line segments are saved in places where fine segmentation is not required, thereby saving the running amount and running time. Finally, the embodiment of the present invention can improve the quality of OPC correction while saving the time for debugging the segmentation rules, and can effectively reduce the error rate of OPC verification.
[0071] The embodiment of the present invention provides an OPC ripple adaptive segmentation algorithm based on a photolithography model, which can improve the OPC correction accuracy and save the segmentation rule adjustment time.
[0072] The OPC ripple dynamic adaptive segmentation algorithm based on the lithography model in the embodiment of the present invention can provide corresponding segmentation solutions according to different lithography models and geometric environments used between different levels in actual production, and can dynamically change the segmentation method and the position allocation of sampling points in multiple iterations.
[0073] The embodiment of the present invention is based on OPC RET technology. In order to address the shortcomings of the existing method for segmenting polygonal layout graphics, such as long time consumption and large amount of computational data, a dynamic adaptive segmentation algorithm based on a lithography model is adopted, so that smaller line segments can be cut out in key areas and the line segments meet the minimum size requirements of layout manufacturing, while long line segments are preserved in places where finer segmentation is not required, saving time for debugging segmentation rules, while improving the quality of OPC correction and effectively reducing the error rate of OPC Verification.
[0074] like Figure 4 , which is the segmentation result of the layout of 4 iterations of the OPC segmentation method based on the lithography model according to the embodiment of the present invention; Figure 4 1, 2, 3 and 4 in the figure correspond to the 1st to 4th iterations respectively. In the 4 iterations, the graphics 102 to be segmented and optimized are all the same, and the segmented graphics are graphics 1021, 1022, 1023 and 1024 respectively. It can be seen that long line segments are saved in places where finer segmentation is not required, which can save running volume and running time; and in key layout areas such as corner areas and line end areas, there are more segmented line segments, which can improve the quality of OPC correction.
[0075] like Figure 5 , which is the simulation result of the layout of 4 iterations of the OPC segmentation method based on the lithography model according to the embodiment of the present invention; Figure 5 1, 2, 3 and 4 correspond to the 1st to 4th iterations respectively. In the 4 iterations, the graphics 102 to be segmented and optimized are all the same, and the simulation graphs corresponding to the segmented graphics 1021, 1022, 1023 and 1024 are graphics 1031, 1032, 1033 and 1034 respectively. It can be seen that the ripples of graphics 1031, 1032, 1033 and 1034 will gradually decrease, so the embodiment of the present invention can realize OPC ripple adaptive segmentation.
[0076] like Figure 6As shown, it is a comparison curve of the number of cuts of two layouts by the OPC cutting method based on the lithography model in the embodiment of the present invention and the existing method; straight line 201a is the number of cuts of the first layout by the existing method, and curve 202a is the number of cuts of the first layout by the OPC cutting method based on the lithography model in the embodiment of the present invention. The value of curve 202a will be reduced, which will save running time. Straight line 201b is the number of cuts of the second layout by the existing method, and curve 202b is the number of cuts of the second layout by the OPC cutting method based on the lithography model in the embodiment of the present invention. The value of curve 202b will be reduced, which will save running time.
[0077] like Figure 7 As shown, it is a comparative bar chart of the number of OPCV errors after the OPC slicing method based on the lithography model of the embodiment of the present invention and the existing method for slicing two layouts; column 203a is the OPCV inspection result after the slicing of the first layout by the existing method, that is, the number of errors, and column 204a is the OPCV inspection result after the slicing of the first layout by the OPC slicing method based on the lithography model of the embodiment of the present invention. The value of curve 204a will decrease, which will improve the quality of OPC correction. Column 203b is the OPCV inspection result after the slicing of the second layout by the existing method, that is, the number of errors, and column 204b is the OPCV inspection result after the slicing of the second layout by the OPC slicing method based on the lithography model of the embodiment of the present invention. The value of curve 204b will decrease, which will improve the quality of OPC correction.
[0078] The present invention has been described in detail above through specific embodiments, but these do not constitute limitations of the present invention. Without departing from the principle of the present invention, those skilled in the art may also make many variations and improvements, which should also be considered as the protection scope of the present invention.
Claims
1. An OPC cutting method based on a photolithography model, characterized in that: The steps include: Step 1: Read the position information of the optimized edge of the graphics to be segmented and optimized in the initial layout, and use the lithography model to simulate the graphics to be segmented and optimized to generate the corresponding contour information of the optimized edge; Step 2: Based on the contour information, a first correction is performed on the overall movement of each optimized edge of the graphics to be segmented and optimized, and a cost function related to light intensity is introduced; The correction order in the first correction is: first correct the long side and then correct the short side, so that the points with the one-dimensional gradient of the cost function greater than 0 and the one-dimensional gradient less than 0 can all appear in the sampling points, so as to ensure that effective segmentation can be performed later; Step 3: Select an initial sampling frequency to sample the light intensity value of the optimized edge, calculate the negative value of the gradient of the cost function at each sampling point, and use interpolation to obtain the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and the troughs; Step 4: based on the points where the gradient of the optimized edge is equal to 0, the positions of the peaks and the troughs, the optimized edge is segmented and the sampling points are adjusted, so that the adjusted sampling points are closer to the peaks or the troughs than the sampling points before the adjustment; Step 5: extract the adjusted light intensity value at each sampling point and calculate the corresponding offset of each optimized segment after optimization; Step 6: Repeat steps 3 to 5 until convergence requirements are met.
2. The OPC cutting method based on the lithography model according to claim 1, characterized in that: In step 1, after the contour information is generated, the step further includes using OPC software to extract all spatial light intensity curves of the graphics to be segmented and optimized and observing the light intensity distribution of all sampling points on the spatial light intensity curves.
3. The OPC cutting method based on lithography model as claimed in claim 2, characterized in that: In step 2, the formula of the cost function is: Where C represents the cost function, r i represents the i-th sampling point, i represents the number of the sampling point, n represents the maximum value of i, I represents the light intensity value, I0 represents the average light intensity value of the optimized edge, I(r i ) represents r i The light intensity value at The gradient formula of the cost function along the normal direction of the i-th sampling point is: in, Represented as r i The normal direction corresponding to the line segment.
4. The OPC cutting method based on lithography model as claimed in claim 2, characterized in that: In step 4, the segmentation includes cutting the optimized edge into a plurality of segmented sections and moving each of the segmented sections to achieve a second correction.
5. The OPC cutting method based on lithography model according to claim 1, characterized in that: The first calibration is performed using OPC software.
6. The OPC cutting method based on lithography model as claimed in claim 4, characterized in that: The second calibration is implemented using OPC software.
7. The OPC cutting method based on lithography model according to claim 1, characterized in that: The initial layout includes multiple layers; In step one, the lithography models used by the initial layouts at different levels are independently set according to the levels.
8. The OPC cutting method based on lithography model according to claim 1, characterized in that: In step 4, in the process of adjusting the sampling points, the sampling points are also adjusted according to the geometric environment.
9. The OPC cutting method based on photolithography model as claimed in claim 8, characterized in that: Adjusting the sampling points according to the geometric environment includes: The number of sampling points is reduced in non-critical layout areas to reduce the amount of calculation and save running time.
10. The OPC cutting method based on lithography model according to claim 8, characterized in that: Adjusting the sampling points according to the geometric environment also includes: In the critical layout area, it is ensured that the number of sampling points can meet the requirements of OPC correction quality.
11. The OPC cutting method based on photolithography model according to claim 10, characterized in that: The key layout areas include: corner areas and line end areas.
12. The OPC cutting method based on photolithography model according to claim 1, characterized in that: In step one, the initial layout is read from the GDS layout file.
13. The OPC cutting method based on lithography model according to claim 1, characterized in that: In step six, satisfying the convergence requirement includes: the offset corresponding to each segment of the optimized edge after optimization is less than or equal to a set value.