Repeater Defect Detection via Hot Scan Coordinate Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor wafer inspection systems face challenges in efficiently detecting repeater defects due to high nuisance levels and throughput limitations, especially with the increasing complexity of feature sizes and the lack of actinic light mask inspectors for EUV lithography, leading to tool choking and dropped defects.
Innovation Solution
A method and system for detecting repeater defects using a hot scan inspection of the entire wafer surface, storing defect positions, determining coordinates for repeater defect detection, and analyzing them using a processor, which includes calculating difference images and filtering out systematic defects to identify defects of interest, utilizing a storage medium like a patch-based virtual inspector or local storage for efficient data management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hot scan inspection is performed with aggressive detection settings to detect all defect types, then measurement precision is improved, but device complexity increases and tool choking occurs due to millions of defect candidates
Solution Approach 1:
The inspection process is divided into two distinct stages: first, a hot scan inspection collects all defect candidates with aggressive settings; second, a repeater defect detection algorithm processes only the coordinate data to identify repeating patterns. This segmentation allows the system to maintain high sensitivity while reducing the burden on the inspection tool by processing defect coordinates rather than full defect data.
Solution Approach 2:
The patent extracts only the essential information (defect coordinates) from the hot scan inspection results, separating this critical data from the full defect information. This extraction enables the repeater detection algorithm to focus on spatial patterns without being overwhelmed by the complete defect dataset, thereby reducing tool complexity while maintaining detection precision.
2Manufacturing precision
If repeater defect detection is performed as the last inspection step with all defects collected, then manufacturing precision is improved, but productivity decreases due to tool choking and dropped defects
Solution Approach 1:
The hot scan inspection is performed first to collect all defect candidates and their coordinates before the repeater defect detection is executed. This preliminary action ensures that all potential repeaters are captured in the initial scan, allowing the subsequent repeater detection algorithm to work with complete coordinate data without compromising throughput, as the heavy lifting of data collection is already done.
Solution Approach 2:
The patent changes the parameter being processed from full defect information to defect coordinates only. This parameter change reduces the data volume and complexity for the repeater detection algorithm, enabling it to process millions of defect candidates efficiently without causing tool choking, thereby maintaining both precision and productivity.
3Measurement precision
If electron beam inspection tool is used for print check with multiple beam/column options, then measurement precision is improved, but productivity decreases with inspection time exceeding 8 hours per reticle
Solution Approach 1:
The repeater defect detection algorithm is designed to be tool-agnostic and can process defect coordinate data from any inspection tool type (electron beam, optical, plasma-based). This universality allows the system to leverage the high precision of electron beam tools when needed while also being compatible with faster optical tools, enabling flexibility in balancing precision and throughput requirements.
Solution Approach 2:
Instead of directly inspecting each reticle with the slow electron beam tool, the system uses defect coordinate data (a copy of the essential inspection information) from a hot scan to identify repeaters. This copying approach allows rapid analysis of defect patterns without requiring time-consuming direct electron beam inspection of every reticle, dramatically improving throughput while maintaining detection accuracy.
4Productivity
If broad band plasma tool is used with higher throughput, then productivity is improved, but measurement precision decreases due to architecture limitations handling less than 10 million defects
Solution Approach 1:
The patent changes the data representation from full defect information to compact coordinate data. This parameter change reduces the data volume to a manageable level that fits within the architecture limitations of plasma-based tools (less than 10 million defects), while still preserving all necessary information for accurate repeater detection, thus maintaining both throughput and precision.
Solution Approach 2:
The system extracts only the essential coordinate information from defect detection, separating this critical data from complete defect descriptions. This extraction enables plasma-based tools with throughput advantages to process defect data efficiently within their architectural constraints while maintaining the ability to detect repeaters accurately through coordinate-based pattern recognition.
Data Source
AI summary
Defects from a hot scan can be saved, such as on persistent storage, random access memory, or a split database. The persistent storage can be patch-based virtual inspector virtual analyzer (VIVA) or local storage. Repeater defect detection jobs can determined and the wafer can be inspected based on the repeater defect detection jobs. Repeater defects can be analyzed and corresponding defect records to the repeater defects can be read from the persistent storage. These results may be returned to the high level defect detection controller.