Multi-Layer Defect Criticality Analysis via Pattern Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pattern-defect analysis methods fail to accurately determine the criticality of defects on a substrate surface due to lack of consideration for relative position relationships between defects and circuit components across different layers, leading to unreliable impact assessments on integrated-circuit functionality.
Innovation Solution
The pattern-defect analysis equipment and method involve acquiring surface observation images, aligning them with design data from multiple layers, and generating synthesized images to determine the relative position relationship between defects and circuit components across different layers, enabling accurate criticality judgments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional defect inspection equipment is used to detect defects based on position coordinate and size data, then detection efficiency is improved, but the ability to analyze relative position relationships between defects and circuit components across different layers deteriorates
Solution Approach 1:
The patent combines defect inspection equipment with defect review equipment into an integrated system. The defect inspection unit detects defects efficiently and provides position data, while the defect review unit performs detailed analysis including multi-layer relative position relationship analysis. The synthesis unit integrates data from both units to provide comprehensive defect criticality assessment, merging the strengths of both equipment types to resolve the contradiction between detection efficiency and information completeness.
Solution Approach 2:
The integrated system performs multiple functions: defect detection, defect classification, multi-layer pattern alignment, relative position relationship analysis, and criticality judgment. By making the system universal and multi-functional, it can handle both high-speed detection and detailed analytical tasks that were previously required from separate equipment, thereby preventing information loss while maintaining detection efficiency.
2Measurement precision
If defect review equipment is used to acquire detailed defect information by magnifying images, then defect type identification is improved, but the ability to judge impact on integrated circuit functions deteriorates
Solution Approach 1:
The patent introduces a new dimension of analysis by performing multi-layer pattern alignment and superimposition. Instead of analyzing defects in isolation or only within their own layer, the system projects and aligns patterns from multiple layers to determine spatial relationships in three-dimensional space. This dimensional extension enables accurate criticality judgment by showing how defects relate to circuit components across different layers, thereby resolving the contradiction between detailed analysis and reliable criticality assessment.
Solution Approach 2:
The synthesis unit acts as an intermediary that receives detailed defect information from the defect review unit and combines it with multi-layer pattern data. It performs the critical function of integrating defect details with spatial context from multiple layers, enabling the system to translate detailed defect information into reliable criticality judgments by mediating between the two types of data.
3Device complexity
If conventional methods analyze only the defect itself without considering relative position relationships, then defect type identification is simplified, but the ability to assess impact on circuit components deteriorates
Solution Approach 1:
The patent segments the analysis process into distinct functional units: defect inspection unit for initial detection, defect review unit for detailed classification, and synthesis unit for integrated analysis. Each unit handles a specific aspect of the analysis, breaking down the complex task of multi-layer relative position relationship analysis into manageable segments. This segmentation maintains analytical simplicity at each stage while collectively achieving comprehensive defect impact assessment.
Data Source
AI summary
A data processing unit acquires a review image including a pattern defect on a substrate, compares the review image with a reference image thereby to extract a defect image, the reference image including no pattern defect, and performs an alignment between the review image and a self-layer design pattern image which is generated from design data belonging to the identical layer in a region corresponding to the review image. The data processing unit, then, based on result of the alignment, generates an another-layer design pattern image which is generated from design data belonging to another layer in the region corresponding to the review image, and, based on a synthesized image of the defect image and the another-layer design pattern image, determines the relative position relationship between the pattern defect and a pattern belonging to another layer, and judges the criticality based on the relative position relationship.