Mask Pattern Layout Partitioning for Consistent OPC Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lithography technologies face challenges in achieving consistent and accurate mask pattern generation due to grid dependency errors, which lead to inconsistent optical proximity corrections (OPCs) and variations in mask patterns for the same target features, especially in inverse lithography techniques like CTM and CTM+, and machine learning-based methods.
Innovation Solution
A method involving partitioning a design layout into cells with defined relationships and assigning variables within these cells to determine a mask pattern that ensures a desired performance metric, using a second coordinate system to minimize grid dependency errors, and iteratively adjusting variables to generate consistent mask patterns across the layout.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If inverse lithography techniques (CTM, CTM+) or machine learning-based methods are used to generate mask patterns, then manufacturing precision and optical proximity correction accuracy are improved, but grid dependency errors cause inconsistency in mask patterns for the same target features
Solution Approach 1:
The design layout is partitioned into multiple overlapping tiles, where each tile is processed independently to generate local mask patterns. This segmentation allows consistent processing of identical target features across different regions while maintaining overall pattern consistency through the overlapping tile structure.
Solution Approach 2:
Multiple local mask patterns generated from overlapping tiles are merged into a single consistent global mask pattern. The merging process resolves conflicts and ensures that identical target features receive identical OPC corrections regardless of which tile they appear in, eliminating grid dependency errors.
2Manufacturing precision
If optical proximity corrections are applied to improve patterning accuracy, then manufacturing precision is improved, but grid dependency introduces variations and inconsistencies in the corrections
Solution Approach 1:
The design layout is pre-processed by partitioning into overlapping tiles before OPC application. This preliminary structuring ensures that identical target features are treated consistently across the entire layout, preventing grid dependency errors from introducing variations in OPC corrections.
Solution Approach 2:
The system evaluates the performance metric of pattern consistency across all tiles and iteratively adjusts the OPC parameters. When inconsistencies are detected in the merged mask pattern, the system refines the local patterns and re-merges them, continuously improving correction consistency until the performance metric is satisfied.
Data Source
AI summary
A method of determining a mask pattern for a target pattern to be printed on a substrate. The method includes partitioning a portion of a design layout including the target pattern into a plurality of cells with reference to a given location on the target pattern; assigning a plurality of variables within a particular cell of the plurality of cells, the particular cell including the target pattern or a portion thereof; and determining, based on values of the plurality of variables, the mask pattern for the target pattern such that a performance metric of a patterning process utilizing the mask pattern is within a desired performance range.