Lithography Feature Selection via Sensitivity Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for selecting features for metrology analysis in integrated circuit lithography are inefficient, generating numerous potential features with limited information on which are the most valuable, leading to excessive and costly analysis, and delayed detection of process deviations.
Innovation Solution
A method that identifies candidate features, defines control regions, performs substrate-level analysis with varying lithography settings, determines feature sensitivity, and selects the most changed features for prioritized metrology analysis, using both reticle design and actual reticle data to focus on the most fragile features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If optical proximity compensation verification is used to identify candidate features, then a comprehensive list of potential fragile features is generated, but the number of features to analyze becomes excessively large with limited guidance on which are most valuable
Solution Approach 1:
The patent extracts only the most critical subset of features from the comprehensive list generated by optical proximity compensation verification. By applying additional filtering criteria and ranking mechanisms, the method extracts the top N most fragile features that provide the highest value for metrology analysis, eliminating the need to analyze all candidate features.
Solution Approach 2:
The patent applies different levels of analysis and selection criteria to different regions or types of features. Instead of uniform treatment, the method identifies specific local characteristics that indicate high fragility and prioritizes those features for detailed metrology analysis, while reducing or eliminating analysis of less critical features.
2Reliability
If more features are selected for analysis to achieve reasonable statistical probability, then the likelihood of selecting fragile features increases, but the time and cost of metrology analysis increases significantly
Solution Approach 1:
The patent changes the selection parameters from random or uniform sampling to a ranked selection based on fragility metrics. By calculating and ranking features according to their fragility scores derived from optical proximity compensation verification and substrate-level analysis, the method achieves high statistical probability of selecting fragile features while analyzing only a small number of top-ranked features.
Solution Approach 2:
The patent performs preliminary ranking and selection of features based on fragility criteria before the actual metrology analysis. This preliminary action identifies and prioritizes the most fragile features in advance, ensuring that the subsequent metrology analysis focuses only on the highest-value targets, thereby reducing overall analysis time while maintaining high detection probability.
3Ease of operation
If uniform sampling of features is used for metrology analysis, then the process is simple to implement, but the most fragile features are not reliably identified
Solution Approach 1:
The patent implements a feedback mechanism where the results of optical proximity compensation verification and substrate-level analysis are used to refine and prioritize the feature selection. The fragility metrics calculated from these analyses provide feedback that guides the selection of the most critical features, ensuring that metrology analysis focuses on features with the highest probability of indicating process deviations.
Solution Approach 2:
The patent performs preliminary analysis using optical proximity compensation verification and substrate-level simulations to identify and rank fragile features before the actual metrology measurement. This preliminary action prepares a prioritized list of features, making the subsequent metrology analysis both simple to implement (by following the pre-determined list) and highly effective (by focusing on the most fragile features).
Data Source
AI summary
A method for selecting a set of features for monitoring a lithography process using a reticle, by identifying a set of candidate features, defining control regions around the candidate features, performing substrate level analysis using the reticle with different settings for the lithography process, determining which of the candidate features are most changed by the different settings, ranking the candidate features according to how much they are changed, and selecting the test set of features from those candidate features that are most changed.