Photomask Inspection Alignment Using Multi-Stage Pattern Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automatic alignment methods for photomask inspection are limited to masks with specific patterns, preventing alignment for general masks without such patterns, thus lacking versatility.
Innovation Solution
An inspection method and apparatus that perform rough and fine alignments using pre-set patterns to rotate a stage with an XY plane, allowing alignment without specific patterns by confirming and correcting position misalignment using optical images of first, second, and third patterns positioned on the mask.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional automatic alignment method using specific patterns is used, then alignment accuracy is improved, but versatility deteriorates because it cannot be applied to general masks without specific patterns
Solution Approach 1:
The patent applies universality by enabling the alignment system to handle both masks with specific patterns and general masks without specific patterns. The method achieves this by using multiple types of patterns (first pattern for rough alignment, second pattern for rough alignment verification, and third patterns for fine alignment) that can be universally applied to any mask type, making the alignment system versatile while maintaining accuracy
2Measurement precision
If rough alignment and fine alignment are performed using multiple patterns, then alignment precision is improved, but inspection time increases
Solution Approach 1:
The patent applies segmentation by dividing the alignment process into two distinct stages: rough alignment using first and second patterns to achieve coarse positioning, and fine alignment using third patterns to achieve precise positioning. This segmentation allows each stage to use patterns optimized for its specific purpose, improving overall alignment precision while managing inspection time efficiently
Solution Approach 2:
The patent applies preliminary action by performing rough alignment before fine alignment. The rough alignment stage using first and second patterns prepares the system by achieving coarse positioning, which reduces the search space and makes the subsequent fine alignment stage more efficient. This preliminary action ensures that the fine alignment starts from a favorable initial state, optimizing the overall process
Data Source
AI summary
An inspection method includes a rough alignment process of acquiring optical images of first and second patterns previously set to confirm whether a position misalignment amount in a rotation direction of a sample with respect to an X or Y direction is equal to or smaller than a first acceptable value, and conforming whether the position misalignment amount is equal to or smaller than the first acceptable value on the basis of an acquisition result, and a fine alignment process of acquiring optical images of third patterns positioned on different corners of a rectangular frame constituted of four sides along the X or Y direction on an optical image of the sample and rotating a stage until a position misalignment amount detected based on the optical images of the third patterns becomes equal to or smaller than a second acceptable value being smaller than the first acceptable value.