Assist Feature Placement via Interference Node Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for placing assist features in mask layouts rely on rule-based methodologies, which are complex and costly to maintain, especially when dealing with off-axis illumination and complex geometries, lacking efficient techniques for determining optimal locations of constructive and destructive interference nodes.
Innovation Solution
The system uses an aerial-image intensity model to iteratively select and place assist features by computing image-gradient magnitudes and updating candidate locations, identifying constructive and destructive interference nodes to optimize assist feature placement, thereby eliminating the need for elaborate rule sets and reducing development time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If rule-based methodology is used for placing assist features, then assist features can be placed based on feature width and spacing parameters, but the system becomes complex and costly to maintain, especially for off-axis illumination and complex geometries
Solution Approach 1:
The patent replaces the mechanical rule-based system with a computational electromagnetic field simulation system. Instead of using elaborate rule sets to determine assist feature placement, the system uses aerial image intensity models and gradient calculations to automatically identify optimal locations. This substitution eliminates the need for manual rule creation and maintenance while providing accurate placement for arbitrary geometries and off-axis illumination conditions.
Solution Approach 2:
The patent changes the approach from using fixed geometric parameters (feature width and spacing) to using optical parameters (aerial image intensity and gradient magnitude). By transforming the placement criteria from simple geometric rules to optical field-based metrics, the system adapts automatically to different illumination conditions and geometries without requiring new rule sets.
2Manufacturing precision
If elaborate rule sets are created to handle off-axis illumination and complex geometries, then assist feature placement accuracy improves, but the cost and difficulty of creating and maintaining rules increases significantly
Solution Approach 1:
The patent replaces the complex rule-based system with a computational electromagnetic field simulation system. Instead of using elaborate rule sets to determine assist feature placement, the system uses aerial image intensity models and gradient calculations to automatically identify optimal locations. This substitution eliminates the need for manual rule creation and maintenance while providing accurate placement for arbitrary geometries and off-axis illumination conditions.
Solution Approach 2:
The system performs self-optimization by automatically calculating the aerial image intensity distribution and identifying constructive and destructive interference nodes without human intervention. The gradient-based algorithm autonomously determines optimal assist feature locations based on the optical physics of the specific layout and illumination conditions, eliminating the need for external rule expertise.
3Measurement precision
If gradient-based iterative method is used to identify interference nodes, then assist feature placement accuracy improves, but computational complexity increases
Solution Approach 1:
The patent performs preliminary calculation of the aerial image intensity distribution across the entire layout before identifying specific assist feature locations. By pre-computing the intensity map and its gradients, the system establishes a foundation that enables rapid identification of constructive and destructive interference nodes through simple gradient magnitude analysis, rather than performing complex iterative optimizations for each potential location.
Solution Approach 2:
The patent changes the approach from using fixed geometric parameters (feature width and spacing) to using optical parameters (aerial image intensity and gradient magnitude). By transforming the placement criteria from simple geometric rules to optical field-based metrics, the system adapts automatically to different illumination conditions and geometries without requiring new rule sets.
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 allows for rapid and accurate placement of assist features, improving lithographic performance and reducing the complexity and cost of maintaining rule-based systems, while converging to optimal solutions in a limited number of iterations.
Implementation Method 1
identifying locations of constructive and destructive interference nodes using the magnitude of an image-intensity gradient of an aerial-image
Data Source
AI summary
One embodiment of the present invention provides a system that determines a location in a layout to place an assist feature. During operation, the system receives a layout of an integrated circuit. Next, the system selects an evaluation point in the layout. The system then chooses a candidate location in the layout for placing an assist feature. Next, the system determines the final location in the layout to place an assist feature by, iteratively, (a) selecting perturbation locations for placing representative assist features in the proximity of the candidate location, (b) computing aerial-images using an image intensity model, the layout, and by placing representative assist features at the candidate location and the perturbation locations, (c) calculating image-gradient magnitudes at the evaluation point based on the aerial-images, and (d) updating the candidate location for the assist feature based on the image-gradient magnitudes.