Defect Detection System Iterative Recipe Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current defect detection methods in semiconductor manufacturing are prone to errors due to mechanical, electrical, or optical issues, leading to false positives and inability to accurately classify defects, especially in high-resolution and high-speed inspection processes.
Innovation Solution
A defect detection system comprising a processing and memory circuitry that processes inspection images using a first recipe to detect potential defects, selects locations for high-resolution review, and iteratively updates the recipe based on classification results until examination stopping criteria are met, incorporating segmentation and grading to enhance defect detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution inspection is performed to detect defects accurately, then measurement precision is improved, but examination time increases and productivity decreases
Solution Approach 1:
The patent segments the inspection process into multiple stages: initial low-resolution screening to identify potential defect regions, followed by targeted high-resolution examination only of those specific regions. This segmentation allows the system to maintain high overall inspection speed while achieving accurate defect detection where needed, resolving the contradiction between measurement precision and productivity.
Solution Approach 2:
The patent applies partial action by performing high-resolution inspection only on selected regions of interest rather than the entire wafer surface. The system uses low-resolution pre-scan data to identify suspicious areas, then applies high-resolution examination partially only to those areas, thereby maintaining productivity while improving defect detection accuracy in critical regions.
2Reliability
If iterative recipe updating is performed to improve defect classification accuracy, then reliability is improved, but examination time increases
Solution Approach 1:
The patent implements feedback mechanisms where classification results from each inspection iteration are fed back to update the inspection recipe for subsequent iterations. The system learns from previous results, refining its defect detection and classification criteria, which improves reliability over time while the iterative process is managed to control time loss through stopping criteria.
Solution Approach 2:
The patent uses periodic action by performing iterative recipe updates at scheduled intervals or after processing a certain number of samples, rather than continuously after every single inspection. This periodic updating maintains improving reliability while preventing excessive time loss, as the system can operate efficiently with established recipes until update points are reached.
3Adaptability or versatility
If multiple inspection recipes are used to classify different defect types, then defect classification capability is improved, but device complexity increases
Solution Approach 1:
The patent applies universality by designing a single inspection system that can execute multiple different inspection recipes through software control. Rather than requiring separate physical inspection devices for each defect type, the system uses one multi-functional platform that can be reconfigured via recipe selection, thereby improving defect classification capability while avoiding the complexity of multiple dedicated devices.
Solution Approach 2:
The patent implements dynamics by making the inspection system adaptable and reconfigurable through dynamic recipe selection. The system can switch between different inspection parameters, thresholds, and analysis methods based on the specific defect type being examined, allowing high versatility without permanent structural complexity. The recipe-based approach enables dynamic adaptation to different inspection requirements.
Data Source
AI summary
Inspection data that corresponds to potential defects of an object may be received. A first set of locations of first potential defects can be identified. The first set of locations of the first potential defects can be imaged with a review tool to obtain a first set of review images. The first potential defects can be classified based on the first set of review images to obtain first classification results of the first potential defects. An instruction can be determined for the review tool based on the first classification results, the instruction being associated with detecting potential defects. Using the instruction, a second set of locations of second potential defects of the plurality of potential defects to be imaged with the review tool can be identified.