Mask Inspection Apparatus for Local Critical Dimension Error Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pattern inspection methods struggle to accurately detect local critical dimension (CD) errors in reticles due to identical deformations causing decreased signal-to-noise ratios, making it difficult to identify microdefects in miniaturized patterns.
Innovation Solution
A pattern inspection apparatus and method that searches for 'look-alike' patterns around a specific pattern, calculates their dissimilarity, and determines local CD errors by excluding allowable errors and comparing against a threshold, effectively identifying deformations through increased distance between patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If intercomparison inspection scheme is used to detect microdefects, then detection capability for extra-small defects is improved, but signal-to-noise ratio decreases due to identical deformations in local regions
Solution Approach 1:
The patent segments the pattern inspection process into two distinct comparison methods: intercomparison for microdefect detection and direct comparison for local CD error detection. This segmentation allows each method to be optimized for its specific detection target, preventing the signal-to-noise ratio degradation that occurs when using intercomparison alone for all defect types.
Solution Approach 2:
Instead of using intercomparison inspection (comparing similar patterns) for local CD errors, the patent inverts the approach by using direct comparison between optical image and fiducial image. This reversal of the comparison strategy eliminates the signal-to-noise ratio problem caused by identical deformations in intercomparison methods.
2Device complexity
If direct comparison between optical image and fiducial image is used, then local CD error detection is simplified, but detection accuracy for microdefects decreases due to pixel position offsets and image expansion/shrink
Solution Approach 1:
The patent divides the inspection system into two parallel processing paths: one for microdefect detection using intercomparison inspection, and another for local CD error detection using direct comparison. This segmentation allows each path to handle the type of defect it is best suited for, avoiding the trade-off between complexity and precision.
Solution Approach 2:
The patent merges the results from intercomparison inspection and direct comparison into a unified inspection system. By combining the strengths of both methods—intercomparison for microdefects and direct comparison for local CD errors—the system achieves comprehensive defect detection without sacrificing accuracy or excessive complexity.
3Productivity
If pattern size is reduced for higher integration, then device integration is improved, but minimum detectable defect size decreases requiring higher inspection accuracy
Solution Approach 1:
The patent implements a dynamic inspection strategy that adapts the comparison method based on the defect type and pattern characteristics. For miniaturized patterns, the system dynamically selects intercomparison inspection for microdefects and direct comparison for local CD errors, maintaining high detection precision across different integration levels.
Solution Approach 2:
The patent changes the inspection parameters (comparison method, threshold values, processing algorithms) based on the specific inspection needs arising from pattern miniaturization. By adjusting these parameters, the system maintains the ability to detect smaller defects while accommodating higher device integration.
Data Source
AI summary
A mask/reticle pattern inspection apparatus capable of readily detecting local critical dimension (CD) errors of a circuit pattern of a testing workpiece is disclosed. This apparatus includes a search unit for finding a plurality of resembling or “look-alike” adjacent patterns around a specific pattern on the workpiece, which have similarity to the specific pattern. The inspection apparatus also includes a calculation unit for obtaining dissimilarity between the specific pattern and look-alike adjacent pattern, a variation evaluation unit which excludes an allowable error from the dissimilarity to thereby obtain a local CD error criterion value, and a CD error decision unit for determining the presence of a local CD error when the criterion value exceeds a threshold value in case the distance between the specific and look-alike patterns increases. A pattern inspection method is also disclosed.