Sampling Map Selection for Semiconductor Alignment Metrology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor wafer fabrication, existing metrology tools face a challenge in balancing alignment measurement accuracy with throughput, as increasing the number of measurements improves precision but decreases production efficiency.
Innovation Solution
A system and method that define a full sampling map with multiple measurement locations, generate candidate sampling maps as subsets, and select a working sampling map by comparing estimated alignment datasets to a reference dataset, prioritizing the smallest number of alignment estimates exceeding a selected tolerance to optimize measurement sites for accurate alignment estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of alignment measurements is increased, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The patent extracts only the essential measurement locations needed to achieve acceptable alignment measurement precision. By identifying and removing redundant measurement points from the full sampling map, the system maintains sufficient measurement accuracy while reducing the total number of measurements performed, thereby improving throughput without sacrificing critical measurement quality.
2Productivity
If the number of measurement locations is reduced, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent performs preliminary actions by first executing a full sampling map measurement to gather comprehensive alignment data across all measurement locations. This reference dataset is then used to evaluate and select the optimal subset of measurement locations for the working sampling map, ensuring that the reduced set will maintain sufficient measurement precision before actual production measurements are performed.
Solution Approach 2:
The patent implements feedback by using the reference alignment dataset obtained from full sampling measurements to evaluate candidate working sampling maps. The system compares estimated alignment values from reduced sampling maps against the reference data, using this feedback to iteratively refine and select the optimal working sampling map that maintains measurement precision while minimizing the number of measurement locations.
3Measurement precision
If a full sampling map is used, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent applies partial action by performing complete full sampling map measurements only once to create a reference alignment dataset. For subsequent production measurements, the system uses a reduced working sampling map that performs only the essential measurements needed, avoiding the excessive time consumption of repeating full sampling maps while maintaining measurement precision through the optimized subset of locations.
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 enhances alignment accuracy while minimizing the number of measurement sites, thereby improving throughput and reducing alignment errors, ensuring precise alignment without excessive resource utilization.
Implementation Method 1
a measured alignment is based on radiation emanating from the sample collected by the detector
Data Source
AI summary
A system for determining a sample map for alignment measurements includes a metrology tool and a controller. The controller defines a full sampling map including a plurality of measurement locations. The controller directs the metrology tool to measure alignment at each measurement location of the full sampling map for a plurality of samples to generate a reference alignment dataset, generates candidate sampling maps, each being a subset of the full sampling map. The controller may further estimate alignment as a function of location based on the two or more candidate sampling maps at each measurement location of the full sampling map, and determine a working sampling map by comparing the estimated alignment to the reference alignment dataset and selecting the candidate sampling map having a smallest number of alignment estimates exceeding a selected tolerance.


