Semiconductor Mask Pattern Extraction and Defect Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for inspecting and correcting semiconductor masks with assist patterns using die-to-database systems face challenges in accurately distinguishing and matching main patterns from assist patterns, leading to errors in defect inspection and correction, especially due to distortions and the complex shapes of modern semiconductor circuits.
Innovation Solution
A semiconductor mask correcting device and method that extracts and recognizes main patterns from mask data by outlining and specifying assist patterns based on predetermined width, length, and space criteria, allowing for precise extraction and matching with actual mask images, and corrects defects using focused ion beams or deposition processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a die-to-database system is used to inspect mask patterns with assist patterns, then inspection accuracy should improve, but pattern matching errors occur due to inability to distinguish main patterns from assist patterns
Solution Approach 1:
The patent segments the mask pattern data into two distinct categories: main patterns and assist patterns. By separating these pattern types and applying different processing rules to each, the system resolves the matching errors that occur when treating all patterns uniformly. Main patterns are matched against the database with strict criteria, while assist patterns are handled differently to prevent false defect detections.
Solution Approach 2:
The patent introduces an intermediary classification mechanism that acts as a mediator between the raw mask data and the defect inspection process. This intermediary step identifies and categorizes patterns as either main or assist patterns before proceeding with defect detection, thereby preventing matching errors between different pattern types.
2Manufacturing precision
If assist patterns are added to improve transfer characteristics, then circuit fabrication quality improves, but defect inspection becomes more difficult due to pattern complexity
Solution Approach 1:
The patent segments the complex mask patterns into main patterns and assist patterns, allowing the inspection system to process them differently. This segmentation reduces the complexity of defect detection by treating assist patterns as reference elements rather than primary inspection targets, thereby maintaining high circuit fabrication quality while simplifying the inspection process.
Solution Approach 2:
The patent applies different inspection criteria and processing methods to different regions of the mask pattern. Main patterns receive rigorous defect inspection, while assist patterns are processed with different rules appropriate to their function as transfer characteristics enhancement elements. This local differentiation resolves the conflict between maintaining fabrication quality and simplifying inspection.
3Length of moving object
If complex fine patterns are used to decrease semiconductor device size, then device miniaturization is achieved, but pattern matching with database becomes more difficult
Solution Approach 1:
The patent segments fine patterns into main and assist pattern categories, enabling the inspection system to handle complex geometries more effectively. By identifying assist patterns separately, the system can focus matching efforts on main patterns while using assist patterns as contextual references, thereby maintaining matching accuracy even for miniaturized device structures.
Solution Approach 2:
The patent changes the processing parameters for different pattern types. For fine main patterns, strict matching criteria are applied, while for assist patterns, different parameters and tolerance levels are used. This parameter differentiation allows accurate inspection of miniaturized patterns without being confounded by the presence of assist patterns with different geometric characteristics.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables high-accuracy pattern matching and correction, improving the quality of semiconductor masks and reducing errors in the fabrication of microscopic circuits by isolating and correcting main patterns effectively, even with assist patterns present.
Implementation Method 1
the protrusion portion is corrected by a focused ion beam to be cut off (i.e. an etching process)
Implementation Method 2
the concave portion is corrected by gas assist method so that a predetermined material is deposited therein (i.e. a deposition process)
Data Source
AI summary
A semiconductor mask correcting device is provided with an image acquiring unit acquiring a mask image, an extraction unit extracting only a main pattern from the mask data, an inspection unit inspecting a defective portion by comparing the extracted main pattern with a main pattern which is obtained from the mask image after a drawing by matching to each other, and a correction unit correcting the defective portion specified by the inspection unit, wherein the extraction unit includes a recognition section recognizing the main pattern and the assist pattern as a figure, a specification section specifying the assist pattern from figures which is recognized on the basis of a predetermined condition, and a main pattern extracting section extracting as the main pattern a figure other than the assist pattern.