Region-Adjusted Inspection Thresholds for Photomask Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pattern inspection apparatuses face challenges in accurately detecting defects in ultrafine patterns due to the generation of pseudo defects, which leads to increased workload and resource wastage, as they struggle to balance precision and false defect detection, especially with the miniaturization of semiconductor and LCD patterns.
Innovation Solution
A pattern inspection apparatus and method that utilizes an optical image acquiring unit, design image data generating unit, and comparing unit, where region image data is used to adjust determination conditions, allowing for variable inspection threshold values based on specific regions to differentiate between critical and non-critical areas, thereby reducing pseudo defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed inspection threshold is used for the entire pattern area, then the inspection process is simple and fast, but it generates many pseudo defects in critical regions while missing real defects in non-critical regions
Solution Approach 1:
The inspection area is divided into multiple regions based on pattern importance. Critical regions (high precision required) and non-critical regions (moderate precision acceptable) are segmented separately. Different threshold values are applied to each segment, allowing optimized defect detection for each region without requiring high precision across the entire area.
Solution Approach 2:
Different inspection threshold values are assigned to different regions of the pattern. Critical regions use stringent thresholds to detect all defects, while non-critical regions use moderate thresholds to reduce pseudo defects. This local differentiation of inspection quality resolves the contradiction between overall precision and complexity.
2Measurement precision
If a stringent inspection threshold is applied to detect all possible defects, then defect detection precision improves, but the number of pseudo defects increases significantly
Solution Approach 1:
The inspection threshold is made non-uniform across the pattern area. Stringent thresholds are applied only to critical regions where defects must be detected, while moderate thresholds are applied to non-critical regions where pseudo defects would occur. This local quality differentiation eliminates the harmful effect of pseudo defects while maintaining detection precision where needed.
Solution Approach 2:
The pattern is segmented into critical and non-critical regions. By segmenting the inspection process, the system can apply different threshold stringency to each segment, preventing pseudo defects in non-critical regions while maintaining high detection precision in critical regions.
3Object-generated harmful factors
If the inspection area is divided into multiple regions with different thresholds, then pseudo defect reduction is achieved, but the processing time and computational load increase
Solution Approach 1:
The inspection area is segmented into critical and non-critical regions, allowing parallel processing with different threshold values. This segmentation enables the system to process different regions simultaneously with optimized thresholds, reducing overall processing time while maintaining pseudo defect reduction benefits.
Solution Approach 2:
Instead of applying maximum inspection stringency to the entire area (excessive action), the system applies stringent thresholds only to critical regions (partial action). This partial application of high inspection quality reduces the total computational load and processing time while still achieving the primary goal of reducing pseudo defects in critical areas.
4Productivity
If uniform inspection thresholds are used across the entire pattern, then the inspection process is fast and simple, but manufacturing yield decreases due to pseudo defects requiring rework
Solution Approach 1:
Different inspection thresholds are applied locally to different regions. Critical regions use stringent thresholds to ensure high manufacturing yield by detecting all defects, while non-critical regions use moderate thresholds to maintain inspection throughput. This local quality differentiation resolves the contradiction between productivity and reliability.
Solution Approach 2:
The inspection process is segmented into critical and non-critical region processing. By segmenting the pattern area and applying different thresholds to each segment, the system maintains high throughput in non-critical regions while ensuring high yield in critical regions, overall improving manufacturing productivity and reliability.
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 effectively reduces the number of pseudo defects, enhances defect detection precision, and optimizes the inspection process by applying stringent or moderate threshold values depending on the region, improving the efficiency and accuracy of pattern inspection.
Implementation Method 1
Light transmitted through the target plate or reflected by the target plate is focused on a sensor through an optical system
Implementation Method 2
Light transmitted through the target plate or reflected by the target plate is focused on a sensor through an optical system
Data Source
AI summary
A pattern inspection apparatus, including an optical image acquiring unit that acquires optical image data of a target plate to be inspected, the target plate being formed as a pattern. The pattern inspection apparatus also includes a design image data generating unit that generates design image data based on a design pattern serving as a base of pattern formation of the target plate. The pattern inspection apparatus further includes a comparing unit that inputs region image data generated based on information of a region pattern which is input to the pattern inspection apparatus. The information of the region pattern represents a predetermined region and is formed in the same format as that of information of the design pattern. The comparing unit compares the optical image data with the design image data based on the region image data.