Mask Defect Inspection Using Sample-Based Nuisance Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect inspection apparatuses struggle to effectively discriminate between defects of interest (DOIs) and nuisances in mask patterns, leading to increased processing loads due to high nuisance detection, which can be exacerbated by setting conditions for maximum sensitivity.
Innovation Solution
A defect inspection method that utilizes a defect inspection apparatus to generate a reference image based on design data, set defect detection conditions using an index from a second sample's inspection results, and discriminate nuisances based on these conditions, incorporating elements like defect categorization, grouping, sensitivity adjustment, and nuisance filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If defect detection sensitivity is maximized by adjusting inspection conditions, then defect detection capability is improved, but the number of nuisances detected increases
Solution Approach 1:
The system performs preliminary defect inspection on a first mask to collect defect data before inspecting the second mask. This preliminary action enables the establishment of defect detection conditions based on actual defect characteristics, allowing the system to differentiate between DOIs and nuisances more effectively during subsequent inspections.
Solution Approach 2:
The system uses feedback from the inspection results of the first mask to adjust and set appropriate defect detection conditions for the second mask. By analyzing the defects detected in the first mask, the system refines its detection parameters to improve DOI identification while reducing nuisance detection in subsequent inspections.
2Measurement precision
If defect detection conditions are set for maximum sensitivity, then defect detection capability is improved, but processing load increases
Solution Approach 1:
The system performs preliminary defect inspection and analysis on a first mask to establish optimal detection conditions before conducting the actual inspection of the second mask. This preliminary characterization of defect types and patterns enables more efficient subsequent inspections with reduced processing requirements.
Solution Approach 2:
The system dynamically adjusts defect detection parameters based on the specific characteristics observed in the first mask inspection. By changing detection parameters according to actual defect patterns rather than using fixed maximum sensitivity settings, the system achieves effective defect detection with reduced processing load.
3Measurement precision
If multiple defects are detected for comprehensive inspection, then defect detection coverage is improved, but discrimination between DOI and nuisance becomes more difficult
Solution Approach 1:
The inspection process is segmented into distinct phases: first, a preliminary inspection of a first mask to characterize defect types; second, using this characterization to guide the inspection of a second mask with targeted parameters. This segmentation allows comprehensive defect detection while simplifying discrimination by categorizing defects based on patterns identified in the first inspection.
Solution Approach 2:
The inspection results from the first mask serve as an intermediary that mediates between comprehensive defect detection and simplified discrimination. By using the first mask's defect characteristics as a reference framework, the system can efficiently categorize and discriminate defects in the second mask without requiring complex real-time analysis of each detected defect.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces the ratio of nuisances detected while maintaining maximum defect detection sensitivity, aligning with user-specific discrimination criteria and reducing processing loads.
Implementation Method 1
a defect inspection apparatus that performs defect inspection on a mask by irradiating a sample with illumination light to acquire an optical image
Data Source
AI summary
According to an embodiment, a defect inspection method is performed with a defect inspection apparatus. The defect inspection apparatus is adapted to irradiate a first sample with illumination light to acquire a first sample image, and compare the first sample image to a reference image to inspect a defect. The defect inspection method includes generating the reference image, acquiring the first sample image, setting a defect detection condition using an index based on a result of defect inspection of a second sample different from the first sample, and discriminating a nuisance from a result of comparison between the reference image and the first sample image based on the defect detection condition.