Methods for adjusting data weights during OPC modeling

Abstract

This invention provides a method for adjusting data weights during OPC modeling. The method involves acquiring ADI (Advanced Device Identification) key dimension data of a wafer and generating a first data file based on these dimensions. Multiple measurement points are then formed from the data file, used to create the OPC model. Weighting rules are set for each measurement point. Initial weights are set for each measurement point according to these rules, followed by optical simulation to form the OPC model. The weights of each measurement point are adjusted according to the weighting rules, and a second data file is generated based on the adjusted weights. The OPC model is then iterated using the second data file. This process is repeated until the edge placement error meets the target value. This invention automatically traverses the data file using pre-set weighting rules, modifies the weights, and outputs a new data file for further fitting and optimization. This shortens the modeling time and avoids errors caused by manual adjustments.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and in particular to a method for adjusting data weights during OPC modeling. Background Technology

[0002] Model-Based Optical Proximity Correction (MBOPC) uses software simulation models to calculate the pattern on the wafer surface after photolithography. Then, by modifying the pattern on the photomask, the simulation results of photolithography are improved, thereby effectively reducing the impact of optical proximity effect on imaging quality and improving resolution.

[0003] The foundation of MBOPC technology application is the establishment of an OPC model. The establishment process can be summarized as follows: (1) After measuring the OPC modeling graphics (such as one-dimensional graphics with different CD and pitch combinations and two-dimensional graphics with two or more lines facing each other) by CDSEM, the data such as line width (Line CD), spacing (Space), and end-to-end space are obtained and sorted out to correspond with the graphics on the mask; (2) The layout of the mask used for modeling, the position information of the measurement points, and the sorted CDSEM data are input into the modeling software. By optimizing the parameters in the software (such as Beam focus, Defocus, AcidDiffusion Length, etc.), the line width and spacing data obtained by the software through simulation calculation are made as close as possible to the CDSEM ADI data obtained by actual measurement, thereby achieving the purpose of simulating the lithography process through the software model.

[0004] OPC model accuracy refers to the difference between the line width, spacing, and other data obtained from software simulation and the measured data. The smaller the difference, the higher the model accuracy. Since building an OPC model usually requires collecting and fitting data from thousands or even tens of thousands of measurement points, it is difficult to obtain optimized results by directly fitting with software. It is necessary to assign different weights to the data of each measurement point and adjust the weights accordingly based on the data fitting results to guide the optimization direction of the software, so as to obtain an OPC model with sufficient accuracy.

[0005] The original measurement point weight adjustment process required checking the measurement data file (gauge file) and, based on the measurement point type, fitting error, and other conditions, finding and adjusting the weights of each corresponding measurement point individually. This process was time-consuming and prone to errors. Please refer to [link / reference needed]. Figures 2A to 2F It shows data from an OPC modeling process in the prior art. Figure 2A The diagram shows the point-like distribution of key size errors in the data file after adjusting the default weights according to the prior art of this invention. Figure 2B This is a schematic diagram showing the critical size error histogram of a data file after adjusting the default weights using existing technology. Figure 2C This diagram illustrates the iterative correction process at the online end of an OPC model generated using existing technology. Figure 2D This is a schematic diagram of the simulated exposure profile of an OPC model generated by existing technology during the iterative correction process; Figure 2E This diagram illustrates how the line fragment movement distance and total movement distance change with the number of iterations in each iteration of the OPC model generated using existing technology. Figure 2F This diagram illustrates the change in line edge placement error during each iteration of the OPC model generated using existing technology.

[0006] To address the aforementioned issues, a novel method for adjusting data weights during OPC modeling is needed. Summary of the Invention

[0007] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a method for adjusting data weights during OPC modeling, which solves the problem that the existing technology requires checking the measurement data file and finding and adjusting the weights of corresponding measurement points one by one according to the measurement point type, fitting error and other conditions, which consumes a lot of time and is prone to errors.

[0008] To achieve the above and other related objectives, this invention provides a method for adjusting data weights during OPC modeling, comprising:

[0009] Step 1: Obtain the ADI critical dimension data of the wafer, and generate a first data file based on the ADI critical dimensions;

[0010] Step 2: Based on the data file, multiple measurement points are formed. These measurement points are used to form an OPC model. Weight setting rules are set for each measurement point.

[0011] Step 3: Set the initial weight of each measurement point according to the weight setting rules, and then perform optical simulation to form the OPC model;

[0012] Step 4: Adjust the weight of each measurement point according to the weight setting rules, and form a second data file based on the adjusted weights;

[0013] Step 5: Iterate the OPC model based on the second data file;

[0014] Step 6: Repeat steps 4 and 5 to iterate the OPC model until its edge placement error meets the target value.

[0015] Preferably, in step one, the critical dimension data of the wafer after photolithography are measured using an electron microscope.

[0016] Preferably, the key dimension data in step one includes the line width, spacing, and line end distance of the cutting layer pattern. The cutting layer pattern is located on a mask and is used to transfer onto the wafer to form a mask pattern. Then, the mask pattern is used as a mask to cut the structure to be cut on the wafer.

[0017] Preferably, the cutting layer pattern in step one includes a one-dimensional pattern with different key dimensions and period combinations, a two-dimensional pattern with two or more lines facing each other, and a pattern with a complexity between the one-dimensional pattern and the two-dimensional pattern.

[0018] Preferably, in step two, a fragment corresponding to the outline of the cutting layer graphic is formed based on the ADI key dimension data, and then the measurement point is formed on the fragment.

[0019] Preferably, in step two, a script tool is used to set the weight setting rules for each measurement point.

[0020] Preferably, the weight setting rule in step two includes adjusting the weight of the graphic complexity corresponding to the measurement point.

[0021] Preferably, in step three, a script tool is used to adjust the initial weights based on the graphical complexity corresponding to the measurement points.

[0022] Preferably, the weight setting rule in step two includes adjusting the weights based on the simulation of the OPC model and the edge placement error of the ADI critical dimension data.

[0023] Preferably, in step four, a script tool is used to adjust the weights based on the edge error.

[0024] Preferably, in step four, the simulated contour of the OPC model after exposure is obtained, and the weight of each measurement point is adjusted according to the weight setting rules based on the edge placement error between the simulated contour and the cutting layer graphic. The second data file is then formed based on the adjusted weights.

[0025] Preferably, the cutting layer pattern in step one is a photoresist layer.

[0026] As described above, the method for adjusting data weights during OPC modeling in this invention has the following beneficial effects:

[0027] This invention proposes a script-based method for automatically adjusting data weights. By using pre-set weight setting rules (such as measurement point type, error value, etc.), the method automatically traverses the data file, modifies the weights, and outputs a new data file for the next step of fitting optimization. This shortens the modeling time and avoids errors caused by manual adjustment. Attached Figure Description

[0028] Figure 1 The diagram shows a model-based optical proximity correction diagram of the prior art.

[0029] Figure 2A This is a schematic diagram showing the point-like distribution of critical size errors in a data file after adjusting the default weights using existing technology.

[0030] Figure 2B This is a schematic diagram showing the critical size error histogram of a data file after adjusting the default weights using existing technology.

[0031] Figure 2C This diagram illustrates the iterative correction process at the online end of an OPC model generated using existing technology.

[0032] Figure 2D This is a schematic diagram of the simulated exposure profile of an OPC model generated by existing technology during the iterative correction process;

[0033] Figure 2E This diagram illustrates how the line fragment movement distance and total movement distance change with the number of iterations in each iteration of the OPC model generated using existing technology.

[0034] Figure 2F This diagram illustrates the change in line edge placement error during each iteration of the OPC model generated using existing technology.

[0035] Figure 3 The diagram illustrates a method for adjusting data weights during the OPC modeling process of this invention.

[0036] Figure 4A The diagram shows the point-like distribution of critical size errors in the data file after adjusting the default weights according to an embodiment of the present invention.

[0037] Figure 4B The diagram shows the key size error histogram of the data file after adjusting the default weights according to an embodiment of the present invention.

[0038] Figure 4C This diagram illustrates the iterative correction process at the online end of the OPC model generated in an embodiment of the present invention.

[0039] Figure 4DThe diagram shows a simulated exposure profile of the OPC model generated in an embodiment of the present invention during the iterative correction process.

[0040] Figure 4E The diagram shows the changes in the line fragment movement distance and total movement distance with the number of iterations during each iteration of the OPC model generated in this embodiment of the invention.

[0041] Figure 4F The diagram shows the change in line end edge placement error during each iteration of the OPC model generated in this embodiment of the invention. Detailed Implementation

[0042] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0043] Please see Figure 3 This invention provides a method for adjusting data weights during OPC modeling, comprising:

[0044] Step 1: Obtain the ADI (After Develop Inspection) of the wafer, which is the measurement of critical dimensions (CD) after development. A layer of photoresist material can be spin-coated on the dielectric layer of the wafer. The dielectric layer can be a metal interconnect layer, gate layer, etc. formed on the wafer. After exposure and development, the critical dimension data is measured again, and a first data file is generated based on the ADI critical dimensions.

[0045] In an embodiment of the present invention, in step one, the key dimension data of the wafer after photolithography are measured by an electron microscope.

[0046] In an embodiment of the present invention, the key dimension data in step one includes the line width, spacing, and line end distance of the cutting layer pattern. The cutting layer pattern is located on a mask and is used to transfer onto the wafer to form a mask pattern. Then, the structure to be cut on the wafer is cut using the mask pattern as a mask.

[0047] In the embodiments of the present invention, the cutting layer pattern in step one includes a one-dimensional pattern with different key dimensions and period combinations, a two-dimensional pattern with two or more lines facing each other, and a pattern with a complexity between the one-dimensional pattern and the two-dimensional pattern. For patterns with different complexities, their impact on optical proximity correction will also be different. That is to say, in the prior art, only the weights are manually adjusted, and the edge placement error of the pattern correction will also be different for patterns with different complexities. It is necessary to continue to correct the pattern to ensure accuracy, which will affect the efficiency and accuracy of the correction.

[0048] In an embodiment of the present invention, the cutting layer pattern in step one is a photoresist layer.

[0049] Step two involves forming multiple measurement points based on the data file. These measurement points are used to form the OPC model. Weighting rules are set for each measurement point. For example, the default weight of the initial OPC (Optical Proximity Correction) program is 1. Measurement points located on or above the fragment (the minimum moving line segment in Optical Proximity Correction) move too much during the iteration process, requiring adjustment of the default weight. In subsequent iterations, the weights of the measurement points also need to be adjusted based on the iteration results to adjust the fragment's movement distance. Existing measurement point weight adjustment processes require checking the measurement data file (gaugefile) and, based on the measurement point type, fitting error, and other conditions, finding and adjusting the weights of each corresponding measurement point individually, which consumes a significant amount of time and is prone to errors.

[0050] In an embodiment of the present invention, in step two, a fragment corresponding to the outline of the cutting layer graphic is formed based on the ADI key dimension data, and then the measurement point is formed on the fragment.

[0051] In an embodiment of the present invention, step two uses a script tool to set the weight setting rules for each measurement point, which improves the speed of OPC modeling.

[0052] In an embodiment of the present invention, the weight setting rule in step two includes adjusting the weight of the graphic complexity corresponding to the measurement point, that is, setting different weights for one-dimensional graphics with different key size and period combinations, two-dimensional graphics with two or more lines facing each other, and graphics with graphic complexity between the one-dimensional graphics and the two-dimensional graphics.

[0053] In an embodiment of the present invention, the weight setting rule in step two includes adjusting the weights based on the edge placement error of the simulation of the OPC model and the ADI key dimension data, that is, the adjusted weights must satisfy the edge placement error tending to 0.

[0054] Step 3, please refer to Figure 4A and Figure 4BThe initial weights of each measurement point are set according to the weight setting rules, and then optical simulation is performed to form the OPC model.

[0055] In an embodiment of the present invention, please refer to Figure 4A and Figure 4B In step three, the initial weights are adjusted by a script tool based on the graphical complexity corresponding to the measurement points, which improves the speed and accuracy of OPC modeling and reduces its error compared to existing technologies.

[0056] Step 4: Adjust the weight of each measurement point according to the weight setting rules, and form a second data file based on the adjusted weights, i.e., form a new data file, which will be used as the inspection file for the next step.

[0057] In an embodiment of the present invention, in step four, a script tool is used to adjust the weights based on the edge error.

[0058] In an embodiment of the present invention, in step four, the simulated contour of the OPC model after exposure is obtained. Based on the edge placement error between the simulated contour and the cutting layer graphic, the weight of each measurement point is adjusted using a script tool according to the weight setting rules. The second data file is then formed based on the adjusted weights.

[0059] Step 5: Iterate the OPC model according to the second data file, that is, move the fragment on it with the weighted measurement points;

[0060] Step six: Repeat steps four and five to iterate the OPC model until its edge placement error meets the target value. Figure 4C This diagram illustrates the iterative correction process at the online end of the OPC model generated in an embodiment of the present invention. Figure 4D The diagram shows a simulated exposure profile of the OPC model generated in an embodiment of the present invention during the iterative correction process. Figure 4E The diagram shows the changes in the line fragment movement distance and total movement distance with the number of iterations during each iteration of the OPC model generated in this embodiment of the invention. Figure 4F The diagram shows the change in line edge placement error during each iteration of the OPC model generated in this embodiment of the invention. That is, during the iteration process, the data file is automatically traversed by pre-set weight setting rules (such as measurement point type, error value, etc.), the weights are modified, and a new data file is output for the next step of fitting optimization. This shortens the modeling time and avoids errors caused by manual adjustment.

[0061] It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0062] In summary, this invention proposes a script-based method for automatically adjusting data weights. By automatically traversing data files using pre-defined weight setting rules (such as measurement point type, error value, etc.), the weights are modified, and a new data file is output for further fitting and optimization. This shortens modeling time and avoids errors caused by manual adjustments. Therefore, this invention effectively overcomes the various shortcomings of existing technologies and has high industrial application value.

[0063] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for adjusting data weights during OPC modeling, characterized in that, At least including: Step 1: Obtain the ADI critical dimension data of the wafer, and generate a first data file based on the ADI critical dimensions; Step 2: Based on the data file, multiple measurement points are formed. These measurement points are used to form an OPC model. Weight setting rules are set for each measurement point. Step 3: Set the initial weight of each measurement point according to the weight setting rules, and then perform optical simulation to form the OPC model; Step 4: Adjust the weight of each measurement point according to the weight setting rules, and form a second data file based on the adjusted weights; Step 5: Iterate the OPC model based on the second data file; Step 6: Repeat steps 4 and 5 to iterate the OPC model until its edge placement error meets the target value.

2. The method for adjusting data weights during OPC modeling according to claim 1, characterized in that: In step one, the key dimensions of the wafer after photolithography are measured using an electron microscope.

3. The method for adjusting data weights during OPC modeling according to claim 2, characterized in that: The key dimensional data mentioned in step one includes the line width, spacing, and line end distance of the cutting layer pattern. The cutting layer pattern is located on a mask and is used to transfer onto the wafer to form a mask pattern. Then, the mask pattern is used as a mask to cut the structure to be cut on the wafer.

4. The method for adjusting data weights during OPC modeling according to claim 3, characterized in that: The cutting layer graphics mentioned in step one include one-dimensional graphics with different key dimensions and period combinations, two-dimensional graphics with two or more lines facing each other, and graphics with graphic complexity between the one-dimensional graphics and the two-dimensional graphics.

5. The method for adjusting data weights during OPC modeling according to claim 4, characterized in that: In step two, a fragment corresponding to the outline of the cutting layer is formed based on the ADI key dimension data, and then the measurement points are formed on the fragment.

6. The method for adjusting data weights during OPC modeling according to claim 1, characterized in that: In step two, a script tool is used to set the weight setting rules for each measurement point.

7. The method for adjusting data weights during OPC modeling according to claim 1, characterized in that: The weight setting rules in step two include adjusting the weights for the graphic complexity corresponding to the measurement points.

8. The method for adjusting data weights during OPC modeling according to claim 7, characterized in that: In step three, a script tool is used to adjust the initial weights based on the graphical complexity corresponding to the measurement points.

9. The method for adjusting data weights during OPC modeling according to claim 1, characterized in that: The weight setting rules in step two include adjusting the weights based on the simulation of the OPC model and the edge placement error of the ADI critical dimension data.

10. The method for adjusting data weights during OPC modeling according to claim 9, characterized in that: In step four, a script tool is used to adjust the weights based on the edge error.

11. The method for adjusting data weights during OPC modeling according to claim 3 or 4, characterized in that: In step four, the simulated outline of the OPC model after exposure is obtained. Based on the edge placement error between the simulated outline and the cutting layer graphic, the weight of each measurement point is adjusted using a script tool according to the weight setting rules. The second data file is then formed based on the adjusted weights.

12. The method for adjusting data weights during OPC modeling according to claim 4, characterized in that: The cutting layer pattern mentioned in step one is a photoresist layer.

Images