Semiconductor Conflict Graph Decomposition for Mask Colorability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor fabrication, determining the colorability of features in a conflict graph becomes increasingly complex and time-consuming as the number of features and links increases, especially when using multiple masks for patterning, which complicates the design and manufacturing process.
Innovation Solution
A method is introduced that involves decomposing and partitioning the conflict graph to reduce complexity, using specific processing devices to analyze and modify the graph, iteratively removing nodes with fewer links than a threshold, and applying predetermined partition rules based on the number of masks used, ultimately determining if the graph is colorable and suggesting mask configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple masks are used to pattern features closer than patterning resolution permits, then the manufacturing precision is improved, but the device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the layout pattern into multiple masks (double patterning, triple patterning) where each mask contains a subset of features. This allows features closer than the patterning resolution to be manufactured with higher precision by separating them into different masks, directly resolving the contradiction between improved manufacturing precision and increased process complexity through systematic decomposition of the patterning process.
2Measurement precision
If the number of features and links in a conflict graph increases, then the measurement precision of colorability determination is improved, but the loss of time increases
Solution Approach 1:
The patent segments the conflict graph into multiple sub-graphs based on the mask assignments and feature separations. This segmentation allows the colorability determination to be performed on smaller, more manageable sub-graphs rather than one large complex graph, maintaining accuracy while significantly reducing the time required for analysis.
Solution Approach 2:
The patent performs preliminary actions by pre-assigning features to masks based on spacing criteria and conflict graph analysis before final colorability determination. This preliminary mask assignment simplifies the subsequent colorability check and reduces the computational time required for the complete determination process.
3Manufacturing precision
If standard cells are checked for colorability based on multiple masks, then the manufacturing precision is improved, but the ease of operation decreases
Solution Approach 1:
The patent implements self-service by automatically performing colorability checks on standard cells during the design process. The system autonomously determines whether standard cells are colorable based on the configured number of masks and spacing criteria, eliminating the need for manual verification and maintaining manufacturing precision while improving design ease through automation.
Data Source
AI summary
A method including decomposing a conflict graph based on a number of masked to be used to manufacture a semiconductor device. The method further includes determining whether the decomposed conflict graph is a simplified graph based on a comparison between the decomposed conflict graph and a stored conflict graph. The method further includes determining whether the decomposed conflict graph is colorable based on a number of masks used to pattern the layer of the semiconductor device. The method further includes indicating that the conflict graph is colorable in response to a determination that the decomposed conflict graph is colorable.