Metrology Target Selection via Simulation Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for selecting metrology targets in lithographic processes face challenges in accurately evaluating target quality and discriminating between suitable and unsuitable targets, often relying on substrate CD change thresholding which can lead to the exclusion of potentially effective targets.
Innovation Solution
A method involving lithographic and metrology simulations is employed to evaluate metrology target designs across a process window, discarding those with catastrophic errors and determining metrology performance indicators for those that pass, allowing for a more comprehensive evaluation of target quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If substrate CD change thresholding is used to evaluate metrology target quality, then the evaluation process is simple and fast, but the accuracy and reliability of target selection deteriorates because potentially effective targets are excluded
Solution Approach 1:
The patent performs preliminary lithographic simulations across the entire process window before final target selection. By pre-evaluating targets under various process conditions (focus, dose, overlay variations), the system identifies potentially effective targets that would be excluded by simple CD thresholding, thereby improving selection accuracy without significantly increasing measurement time
Solution Approach 2:
The patent creates virtual copies of metrology targets through simulation models that replicate the lithographic process. These simulated target images allow comprehensive evaluation of target performance across the process window without physically manufacturing and measuring each target variant, enabling thorough quality discrimination while maintaining efficiency
2Measurement precision
If comprehensive simulation evaluation is performed for all process window points, then target quality discrimination accuracy is improved, but the computational time and complexity increases
Solution Approach 1:
The patent segments the process window into critical regions (corners and center) rather than uniformly evaluating all points. By identifying and focusing simulation resources on the most critical process conditions where catastrophic errors are most likely to occur, the system achieves comprehensive quality assessment with reduced computational time
Solution Approach 2:
The patent performs preliminary simulations at process window corners and center points to identify targets that maintain stability across all conditions. This preliminary filtering approach eliminates obviously unsuitable targets before comprehensive evaluation, reducing the total computational burden while maintaining high accuracy in target quality discrimination
3Ease of manufacture
If metrology targets are selected based on CD change thresholding, then the selection process is straightforward, but the reliability of selected targets deteriorates under varying process conditions
Solution Approach 1:
The patent evaluates metrology targets across multiple parameter variations including focus offset, dose variation, and overlay error conditions. By simulating target performance under these changing parameters, the system identifies targets that maintain stable and reliable performance across the process window, rather than selecting targets based on single-point CD measurements that may be sensitive to process variations
Data Source
AI summary
A system to, and a method to, select a metrology target for use on a substrate including performing a lithographic simulation for a plurality of points on a process window region for each proposed target, identifying a catastrophic error for any of the plurality of points for each proposed target, eliminating each target having a catastrophic error at any of the plurality of points, performing a metrology simulation to determine a parameter over the process window for each target not having a catastrophic error at any of the plurality of points, and using the one or more resulting determined simulated parameters to evaluate target quality.