OPC Model Construction Using Measured Mask Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical proximity correction (OPC) models fail to accurately account for deviations in mask patterns formed during the fabrication process, leading to inaccuracies in transferring designs to semiconductor wafers due to corner rounding and other optical proximity effects, which affect critical dimension and yield.
Innovation Solution
A method is introduced to construct an OPC model by measuring the actual pattern formed on the mask using techniques like scanning electron microscopy and inputting physical and statistical parameters, then using this data to create a database and correct the pattern transfer to the photoresist layer, enhancing the accuracy of lithography processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the original design pattern is used to construct the OPC model, then the model construction process is simple, but the pattern transfer accuracy deteriorates due to corner rounding and optical proximity effects
Solution Approach 1:
The patent applies preliminary action by measuring the actual mask pattern before constructing the OPC model. The mask pattern is measured using scanning electron microscopy or other measurement tools to obtain the real pattern geometry, including corner rounding and dimensional changes. This measured actual pattern is then used as the basis for constructing the OPC model, ensuring that the model reflects the true mask pattern rather than the idealized design pattern. This preliminary measurement step resolves the contradiction by sacrificing construction simplicity for improved pattern transfer accuracy.
2Manufacturing precision
If the mask pattern is measured to construct the OPC model, then the pattern transfer accuracy is improved, but the process complexity increases
Solution Approach 1:
The patent introduces an intermediary measurement process that bridges the gap between the original design pattern and the actual mask pattern. The measurement system (such as scanning electron microscopy) serves as an intermediary that captures the actual mask pattern geometry. This measured data then becomes the foundation for constructing the OPC model. By using this intermediary measurement approach, the patent improves pattern transfer accuracy while managing process complexity through standardized measurement procedures and automated data processing.
3Ease of operation
If conventional OPC methods are used, then the process is straightforward, but the yield deteriorates due to unaccounted optical proximity effects
Solution Approach 1:
The patent implements feedback by measuring the actual mask pattern and using this information to construct an updated OPC model. The measured mask pattern data is fed back into the OPC modeling process, allowing the model to account for real-world effects such as corner rounding and optical proximity effects that were not present in the original design. This feedback mechanism improves device yield by ensuring that the OPC corrections are based on actual measured deviations rather than theoretical assumptions, while maintaining operational simplicity through automated model construction.
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
This approach improves the authenticity and reliability of the lithography process by accurately accounting for pattern deviations, reducing errors and enhancing the yield of semiconductor devices by constructing an OPC model based on the actual mask pattern, rather than the original design.
Implementation Method 1
using an image obtained by a scanning electron microscope (SEM)
Data Source
AI summary
A method for constructing an optical proximity correction (OPC) model is described. A test pattern is provided, and the test pattern is then written on a mask. The pattern on the mask is measured to obtain a modified pattern. An OPC model is constructed according to the modified pattern.