Semiconductor Pattern Dimension Measurement Using Multi-Focus Image Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is an increasing demand for a reliable method and system to measure the dimensions of fine patterns in semiconductor devices, which are scaled down, and existing inspection processes are inefficient in terms of time and cost.
Innovation Solution
A method and system that involve performing focusing operations on semiconductor samples by scanning at different focusing levels to obtain multiple focus images, using these images to measure pattern dimensions, and employing a data processing unit to select the best focus images for accurate dimension measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple scanning operations are performed to obtain clear focus images for accurate measurement, then measurement precision is improved, but inspection time increases
Solution Approach 1:
The patent performs a preliminary focusing operation that scans at multiple focusing levels to obtain a plurality of focus images before the actual dimension measurement. This preliminary action captures all necessary image data in advance, so that the subsequent measurement can directly use the clearest image without requiring additional scanning operations, thereby resolving the contradiction between measurement precision and inspection time.
Solution Approach 2:
The patent segments the inspection process into two distinct phases: (1) a focusing operation phase that obtains multiple focus images at different focusing levels, and (2) a measurement phase that selects the clearest image for dimension measurement. This segmentation allows the system to gather comprehensive image data efficiently in the first phase, then perform rapid selection and measurement in the second phase, avoiding repeated scanning.
2Reliability
If comprehensive inspection processes are implemented to ensure reliable measurement of fine patterns, then measurement reliability is improved, but inspection cost increases
Solution Approach 1:
The patent implements a self-service mechanism where the system automatically evaluates the clarity of multiple focus images and selects the most suitable one for measurement without requiring manual intervention or complex additional equipment. The data processing unit autonomously determines which image provides the best measurement reliability, reducing the need for expensive manual inspection processes while maintaining high measurement reliability.
Solution Approach 2:
The patent changes the focusing level parameter during the scanning process to obtain images at multiple focusing levels. By varying this single parameter systematically, the system achieves comprehensive inspection data with a single scanning operation, avoiding the need for multiple expensive scanning operations or complex inspection equipment while ensuring reliable measurement of fine patterns.
3Measurement precision
If additional scanning operations are performed to obtain clear focus images, then measurement accuracy is improved, but device complexity increases
Solution Approach 1:
The patent merges the focusing operation and image acquisition into a single integrated process. The same scanning operation that moves the scanner across the sample also varies the focusing level to capture multiple focus images. This merging eliminates the need for separate focusing adjustments or additional scanning operations, maintaining measurement accuracy while reducing device complexity and operational complexity.
Data Source
AI summary
A method of inspecting a sample includes performing a focusing operation on a target pattern of a sample. The focusing operation includes scanning the target pattern at different focusing levels to obtain a plurality of focus images. The method further includes using at least one of the plurality of the focus images as a target pattern image of the target pattern and then measuring a dimension of the target pattern based on the target pattern image.