Charged Particle Beam Apparatus Memory Cell Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional scanning electron microscopes lack the high-accuracy sample stage mechanism required to identify specific memory cells in miniaturized semiconductor memory devices, making it difficult to accurately count and locate memory cells.
Innovation Solution
A charged particle beam apparatus that generates a scale pattern corresponding to repeated memory cells, allowing for image superimposition and software-based zoom and shift without displacing the sample stage, enabling the identification and counting of memory cells using a combination of beam deflection and software-based image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional scanning electron microscope is used, then the apparatus is simple and easy to operate, but it cannot achieve the required submicron-level positioning accuracy for identifying specific memory cells
Solution Approach 1:
The patent replaces the mechanical sample stage positioning system with a software-based image processing system. Instead of relying on mechanical displacement accuracy, the system uses scale pattern recognition, image superimposition, and coordinate transformation algorithms to achieve submicron-level positioning accuracy through digital processing rather than mechanical precision.
Solution Approach 2:
The patent creates a virtual scale pattern that replicates the periodic structure of memory cells. By generating this digital copy of the cell pattern and using it for image superimposition and correlation analysis, the system can identify specific cell positions without requiring the physical sample stage to achieve the corresponding mechanical precision.
2Measurement precision
If the sample stage is displaced with high precision to identify specific memory cells, then the positioning accuracy is improved, but the device complexity and cost increase significantly
Solution Approach 1:
The patent replaces the mechanical sample stage positioning system with a software-based image processing system. Instead of relying on mechanical displacement accuracy, the system uses scale pattern recognition, image superimposition, and coordinate transformation algorithms to achieve submicron-level positioning accuracy through digital processing rather than mechanical precision.
Solution Approach 2:
The patent creates a virtual scale pattern that replicates the periodic structure of memory cells. By generating this digital copy of the cell pattern and using it for image superimposition and correlation analysis, the system can identify specific cell positions without requiring the physical sample stage to achieve the corresponding mechanical precision.
3Productivity
If software-based zoom and image shift are implemented without sample stage displacement, then the operation speed and efficiency are improved, but the complexity of image processing increases
Solution Approach 1:
The patent performs preliminary actions by pre-generating scale patterns based on the known periodic structure of memory cells and pre-establishing the coordinate transformation relationships between the scale pattern and the actual cell positions. This preparation work enables rapid image superimposition and cell identification during operation, improving efficiency without requiring complex real-time processing.
Solution Approach 2:
The patent employs feedback mechanisms through image correlation analysis and pattern matching. The system compares the superimposed scale pattern with the actual observed cell pattern, uses the correlation results to refine position identification, and provides feedback for iterative optimization, thereby managing processing complexity through intelligent feedback loops.
Data Source
AI summary
When a sample includes repeated cells, a scale pattern corresponding to the repeated cells is generated. Next, the scale pattern generated is superimposed on the image of the repeated cells of the sample, thereby identifying a destination cell. Moreover, disposition of the repeated cells of the sample is determined based on positions of at least three ends of the repeated cells. Then, the position of the destination cell is identified from this disposition of the repeated cells. Furthermore, a zoom image is generated by a combination of a zoom based on beam deflection function and a zoom based on software. Then, the image shift is performed by software without displacing a sample stage.