Lithography Pattern Edge Correction Using Open Area Density
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional loading effect correction methods in semiconductor manufacturing, particularly in optical lithography, are inadequate for accurately adjusting pattern edges due to their reliance on dose-based corrections, which fail to account for the geometric variations caused by neighborhood open area density, leading to non-optimal results, especially with the advent of smaller mask shapes and dose manipulation techniques.
Innovation Solution
Geometric loading effect correction (gLEC) adjusts pattern edges based on calculated offsets using neighborhood open area density, applying individual pixel dose changes to counteract loading effects, applicable to both VSB and multi-beam machines, and suitable for both Manhattan and curvilinear shapes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If dose-based loading effect correction methods are used, then the correction process is simple and straightforward, but the accuracy of pattern edge adjustment deteriorates due to inability to account for geometric variations caused by neighborhood open area density
Solution Approach 1:
The patent changes the correction parameter from dose-based to geometry-based. Specifically, it uses neighborhood open area density (a geometric parameter) to calculate loading effect corrections instead of using dose-based methods. This parameter change enables accurate accounting for geometric variations in pattern density while maintaining computational feasibility through established correction frameworks.
Solution Approach 2:
The patent performs preliminary calculation of neighborhood open area density for each pattern before applying the loading effect correction. By pre-calculating the geometric density parameters and storing them, the system prepares the necessary geometric information in advance, which is then used to determine the appropriate correction amounts without adding complexity to the main correction process.
2Reliability
If conventional loading effect correction methods are used, then the manufacturing process is straightforward, but the reliability of pattern formation deteriorates due to non-optimal results with smaller mask shapes and dose manipulation techniques
Solution Approach 1:
The patent addresses reliability issues with conventional methods by changing from dose-based parameters to geometry-based parameters (neighborhood open area density). This parameter change makes the correction method more reliable for smaller mask shapes and dose manipulation techniques, as geometric density directly reflects the physical loading conditions without being confounded by dose variations.
Solution Approach 2:
The patent implements a feedback mechanism where the calculated loading effect corrections are applied to adjust pattern edges, and this correction process can be iterated if needed. The system uses the neighborhood open area density as feedback information to continuously refine the correction amounts, improving reliability through iterative optimization while maintaining manageable complexity.
3Manufacturing precision
If geometric loading effect correction using neighborhood open area density is implemented, then pattern edge accuracy is improved, but the calculation and processing complexity increases
Solution Approach 1:
The patent segments the correction process into distinct steps: (1) calculating neighborhood open area density for each pattern, (2) determining loading effect correction amounts based on these density values, and (3) applying corrections to pattern edges. This segmentation allows each step to be optimized independently and facilitates parallel processing, reducing overall computational complexity while maintaining high edge adjustment accuracy.
Solution Approach 2:
The patent applies local quality by calculating neighborhood open area density specifically for each pattern's local environment rather than using global averages. This localized approach provides accurate edge adjustment for each pattern based on its specific geometric context, while the modular nature of local calculations allows efficient processing without requiring complex global optimization.
Data Source
AI summary
Methods and systems involve a plurality of patterns, each pattern in the plurality of patterns comprising a plurality of edges. Methods and systems also involve determining a neighborhood open area density for the plurality of patterns; determining a geometric loading effect correction, wherein the geometric loading effect correction comprises a calculated offset from an edge of a pattern in the plurality of patterns, and wherein the calculated offset is determined using the neighborhood open area density; and adjusting the edge of the pattern in the plurality of patterns using the geometric loading effect correction.