Optical Proximity Correction Model Using Pattern Density Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical proximity correction (OPC) modeling methods fail to account for pattern density, leading to inaccuracies in critical dimension (CD) and process window, especially as photolithography approaches its resolution limit.
Innovation Solution
A modeling method and system that configures OPC test patterns with varying pattern densities, including primary and auxiliary patterns, to create a mapping relationship between actual on-wafer CD and pattern density, enabling accurate CD correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional OPC modeling methods are used without considering pattern density, then the modeling process is simple and fast, but the manufacturing precision of critical dimension deteriorates
Solution Approach 1:
The patent introduces pattern density as a new parameter in the OPC model. The model now includes pattern density variables (such as local pattern density within a predetermined region) in addition to traditional parameters. This allows the model to account for the influence of surrounding patterns on the critical dimension, thereby improving manufacturing precision without excessive complexity increase.
Solution Approach 2:
The patent segments the pattern density calculation into local and global components. Local pattern density is calculated within a predetermined region around the target pattern, while global pattern density considers the entire chip. This segmentation allows the model to capture local effects that most influence critical dimension while maintaining manageable computational complexity.
2Manufacturing precision
If pattern density is taken into account in OPC modeling, then the manufacturing precision of critical dimension improves, but the device complexity increases
Solution Approach 1:
The patent implements partial action by focusing pattern density calculation on a predetermined local region around the target pattern rather than the entire chip. This local pattern density approach captures the most significant influences on critical dimension while avoiding the excessive computational complexity of global analysis. The predetermined region is optimized to include only the necessary surrounding patterns that affect the target pattern's critical dimension.
3Reliability
If conventional OPC modeling is used, then the ease of manufacture is maintained, but the reliability of corrected patterns deteriorates
Solution Approach 1:
The patent performs preliminary action by pre-calculating pattern density values and storing them in lookup tables before actual OPC correction. The model pre-establishes the relationship between pattern density and critical dimension variation through calibration wafers. During production, the system simply queries these pre-computed values rather than performing complex real-time calculations, thereby improving reliability while maintaining ease of manufacture.
Data Source
AI summary
A modeling method and a modeling system for an optical proximity correction (OPC) model and an OPC method based on the OPC model are disclosed. During the creation of the OPC model, variables related to pattern density of a test pattern are added. As a result, the created innovative OPC model includes a mapping relationship between a critical dimension of the pattern and the pattern density-related variables. After that, the OPC model is used to determine critical dimension (CD) distortion of the test pattern in environments with different pattern densities, which is then compared with a designed CD for the test pattern, thereby determining an OPC correction amount for the test pattern. In this way, accurate OPC correction can be achieved, resulting in an improved process window for the layout and effectively increased yield of the product.