Defect Observation System Field of View Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increased complexity and miniaturization in semiconductor production lead to more types of defects, necessitating the detection of minute defects, which are often obscured by contamination or local electrification during low-magnification image acquisition, causing errors in defect detection and classification.
Innovation Solution
The field of view for high-magnification images is set to exclude the outer edges of the low-magnification image, preventing contamination or local electrification from influencing the high-magnification image, thereby ensuring accurate defect detection and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a low-magnification image is acquired to detect defect coordinates, then defect detection coverage is improved, but contamination or local electrification is generated at the outer edges affecting image quality
Solution Approach 1:
The image acquisition process is segmented into two distinct stages: low-magnification imaging for defect coordinate detection and high-magnification imaging for detailed observation. By separating these functions and acquiring high-magnification images only of regions excluding the contaminated outer edges, the system eliminates the harmful effects while preserving the benefits of both imaging stages.
Solution Approach 2:
The patent applies local quality by adjusting the field of view parameters specifically for high-magnification imaging to exclude the outer edge portions where contamination occurs. This localized adjustment ensures that only the clean central region is captured at high magnification, maintaining image quality without sacrificing the overall defect detection capability.
2Area of stationary object
If the field of view is increased to capture more defect regions, then defect detection completeness is improved, but the outer edges containing contamination are included in the high-magnification image
Solution Approach 1:
The patent dynamically adjusts the field of view parameters based on the detected defect coordinates and the known contamination pattern. The high-magnification field of view is adaptively positioned and sized to cover defect regions while excluding the contaminated outer edges, optimizing both coverage and reliability for each specific observation task.
3Measurement precision
If multiple frame images are integrated to reduce noise for minute defect detection, then detection sensitivity is improved, but exposure time is increased allowing more contamination to accumulate
Solution Approach 1:
The patent performs preliminary action by acquiring low-magnification images first to detect defect coordinates and determine the optimal high-magnification field of view parameters. This preliminary detection step allows subsequent high-magnification imaging to be focused only on clean regions, reducing the need for extensive frame integration and minimizing total exposure time while maintaining detection sensitivity.
Data Source
AI summary
When contamination or local electrification is generated during acquisition of a low-magnification image, if a high-magnification image contains both a portion in which the contamination or local electrification is generated and a portion in which the contamination or local electrification is not generated, a region whose image quality has changed due to the contamination or local electrification is erroneously recognized as a defect. Thus, defect detection fails or it may be impossible to correctly determine the feature quantity of a defect. The invention provides a defect observation system that acquires sample images at a low magnification and a high magnification, and sets the position or size of the field of view of the high-magnification image or the electron beam irradiation range during acquisition of the low-magnification image no that the image acquired at the high magnification does not contain the outer edge of the image acquired at the low magnification.