Pattern Matching for Multilayer SEM Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor device fabrication, pattern matching for multilayer structures is challenged by significant pattern deformations and variations in secondary electron detection efficiency, leading to decreased measurement accuracy and throughput, especially for contact holes with deep three-dimensional structures.
Innovation Solution
A method and device for pattern matching that uses a first template representing the upper portion of a pattern for initial matching, followed by selective extraction of the lower portion's shape information, allowing for stable positioning and selective information extraction across layers, regardless of depth and charging effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pattern matching is performed on multilayer structures with deep three-dimensional patterns, then positioning information can be obtained, but pattern deformations and secondary electron detection variations cause matching failure and decreased measurement accuracy
Solution Approach 1:
The patent segments the pattern matching process into two distinct stages: first matching the upper layer pattern to obtain positioning information, then separately extracting and matching the lower layer pattern. This segmentation allows each stage to focus on specific layer characteristics, reducing the impact of deformations and detection variations that would affect a single comprehensive matching attempt.
Solution Approach 2:
The patent performs preliminary action by first matching the upper layer pattern before attempting to match the lower layer pattern. This preliminary positioning establishes a reference framework that guides subsequent lower layer extraction, ensuring that matching attempts are focused on the correct regions and reducing failures due to positioning errors.
2Ease of operation
If conventional pattern matching is used on deep hole patterns, then positioning can be attempted, but electrification effects and secondary electron variations cause orbit deviations and detection yield variations
Solution Approach 1:
The patent separates the positioning task into two segments: upper layer positioning using upper layer pattern matching, and lower layer positioning using lower layer pattern extraction. This segmentation allows each positioning task to be optimized for its specific layer's characteristics, reducing the impact of electrification effects and secondary electron variations that affect deep hole patterns.
3Loss of information
If the lower portion of hole patterns is matched directly, then deep pattern information can be obtained, but significant deformations and charging effects reduce matching success rate
Solution Approach 1:
The patent performs preliminary action by first matching the upper layer pattern to establish positioning information before extracting the lower layer pattern. This preliminary step provides a reference framework that guides the subsequent lower layer extraction, ensuring that information is obtained reliably even when direct matching would fail due to deformations and charging effects.
Solution Approach 2:
The patent extracts the lower layer pattern information separately after establishing upper layer positioning, rather than attempting to match the entire deep pattern at once. This extraction approach isolates the lower layer information retrieval from the complications of deep pattern deformations and charging effects, improving matching success rate.
Data Source
AI summary
A pattern matching method for a scanning electron microscope comprises a step of performing pattern matching of only an upper layer pattern between an image (101) in which a pattern consisting of plural layers is represented and a template (104) in which the upper layer pattern of the plural layer pattern is selectively represented, thereby identifying the position of the pattern consisting of the plural layers. Then, information about the upper layer pattern is subtracted from the image (101), thus extracting shape information (108) about the lower layer pattern. Consequently, stable positioning or selective information extraction on a certain layer is enabled regardless of the state of the depths of a pattern formed in three dimensions or of the charge state of a sample.