Block-to-block reticle inspection for single die patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current reticle inspection methods, such as die-to-die, die-to-database, and cell-to-cell inspections, are inadequate for reticles with single dies or dies of different designs, and are impractical due to data flow infrastructure limitations and large database sizes, especially when comparing distant portions of the reticle.
Innovation Solution
A method and system for block-to-block reticle inspection that acquires swath images, identifies and aligns repeating blocks, determines geometrical characteristics and spatial offsets, and generates difference images to detect defects, either with or without design database assistance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If die-to-die inspection is used, then inspection coverage is improved, but it becomes impossible when reticle contains single die or dies of different designs
Solution Approach 1:
The inspection method segments the reticle into multiple blocks and performs block-to-block comparison. This allows inspection of repeating patterns even when the entire reticle contains only a single die or dies of different designs, by focusing on localized repeating blocks rather than requiring complete die-to-die matching
Solution Approach 2:
The method creates a reference block from one location and compares it against other occurrences of the same block type. This copying approach enables inspection of repeating blocks without requiring the entire die or reticle to be identical, solving the limitation of die-to-die inspection for single die or different design scenarios
2Manufacturing precision
If die-to-database inspection is used, then inspection thoroughness is improved, but it becomes impractical due to data flow infrastructure limitations and large database sizes
Solution Approach 1:
The method extracts only the necessary block pattern information from the reticle design data and stores it in a compact format. This extraction approach eliminates the need for large complete reticle databases and complex data flow infrastructure, while maintaining inspection thoroughness by focusing on the essential repeating block characteristics
Solution Approach 2:
Instead of maintaining large, complex design databases, the method uses lightweight block pattern references that can be generated on-demand from the acquired image data. This disposable approach to pattern reference eliminates the need for persistent large-scale database infrastructure
3Measurement precision
If cell-to-cell inspection is used, then local pattern matching is improved, but it is limited when comparing portions separated by large distances
Solution Approach 1:
The method extends the inspection capability from local cell-to-cell comparison to block-to-block comparison across the entire reticle by using coordinate transformation and spatial offset calculation. This dimensional extension allows comparison of blocks separated by large distances while maintaining the precision of local pattern matching through mathematical coordinate mapping
4Ease of manufacture
If traditional inspection methods are used, then processing simplicity is maintained, but inspection effectiveness for complex reticle designs with large spatial separations deteriorates
Solution Approach 1:
The complex reticle inspection problem is segmented into manageable blocks that can be independently analyzed and compared. This segmentation maintains processing simplicity by breaking down complex patterns into smaller units while improving inspection effectiveness for complex designs with large spatial separations between repeating elements
Solution Approach 2:
The method introduces spatial coordinate transformation and block mapping dimensions to handle complex reticle layouts. This allows the system to maintain simple block-to-block comparison logic while effectively inspecting complex designs with large spatial separations through coordinate-based block identification and matching
Data Source
AI summary
Block-to-block reticle inspection includes acquiring a swath image of a portion of a reticle with a reticle inspection sub-system, identifying a first occurrence of a block in the swatch image and at least a second occurrence of the block in the swath image substantially similar to the first occurrence of the block and determining at least one of a location, one or more geometrical characteristics of the block and a spatial offset between the first occurrence of the block and the at least a second occurrence of the block.