Semiconductor Overlay Control via Calibration Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face challenges in accurately determining and compensating for overlay errors, particularly at smaller scales (less than hundreds of micrometers), which can lead to malfunctions such as connection failures and short circuits due to misalignment of layers.
Innovation Solution
A method and system for overlay control in semiconductor devices that involves obtaining data from a selected area of the semiconductor device pattern, calculating overlay errors, and generating a calibration image to create a calibrated area, which can then be used to generate a photomask for improved layer alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current metrology equipment is used to measure overlay errors, then measurement can be performed on larger scales, but measurement precision deteriorates at smaller scales (less than hundreds of micrometers)
Solution Approach 1:
The patent introduces a calibration image as an intermediary reference object between the metrology equipment and the actual semiconductor device patterns. This calibration image contains known dimensional references that enable the measurement system to accurately determine overlay errors at small scales by comparing measured features against the calibration standards, effectively bridging the gap between the equipment's native measurement capabilities and the required small-scale precision.
Solution Approach 2:
The patent changes the measurement parameters by introducing calibration data with known dimensions and using these to adjust and correct the measurement scale. By incorporating calibration images with predetermined features at specific scales, the system transforms raw measurements into accurate overlay error values even at dimensions below the equipment's native resolution limits.
2Manufacturing precision
If overlay control is not performed accurately, then manufacturing process is simpler, but manufacturing precision deteriorates due to layer misalignment
Solution Approach 1:
The patent performs preliminary calibration by measuring features in the calibration image before measuring the actual device patterns. This preliminary action establishes reference values and correction factors that are then applied to subsequent measurements, simplifying the overall process by pre-establishing the measurement baseline and reducing the complexity of real-time overlay control.
Solution Approach 2:
The patent uses a calibration image as a copy or replica of known reference structures. By measuring this copied reference pattern and comparing it against the original known dimensions, the system establishes accurate measurement parameters that can then be applied to the actual device patterns, reducing the complexity of direct measurement while maintaining high precision.
3Manufacturing precision
If calibration images are generated with accurate dimensions, then overlay error compensation improves, but device complexity increases due to additional calibration steps
Solution Approach 1:
The patent merges the calibration information directly into the photomask data structure by generating calibration images with known dimensions and integrating their measurement results into the overlay control workflow. This merging allows the calibration process to become part of the standard photomask generation flow rather than a separate add-on, reducing overall process complexity while maintaining accurate overlay error compensation.
Data Source
AI summary
A method for overlay control based on a semiconductor device pattern includes: obtaining data of selected area from the semiconductor device pattern that includes a wiring section with a first pattern image and a second pattern image that are images of two wiring patterns formed in different steps in a semiconductor fabrication process; obtaining two overlay errors associated with the first pattern image and the second pattern image on two different locations; calculating a calibration dimension based on either the overlay errors or a pre-set value; and generating a calibrated area based on the selected area and the calibration dimension.