Overlay Metrology Sampling Plans for Out-of-Spec Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for overlay metrology in semiconductor manufacturing often miss out-of-specification lots due to random selection, leading to reduced yield and potential false detections, as they fail to separate effectively between run-to-run control and out-of-specification detection, and do not optimize metrology tool settings and sampling for higher throughput.
Innovation Solution
A system and method for generating sparse sampling plans that focus specifically on out-of-specification detection, determining the number and locations of metrology points, and optimizing metrology tool settings to maximize detection while minimizing throughput impact, using a computer system to generate sampling plans and perform metrology processes with reduced measurement points per wafer or lot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If random selection of wafers for metrology is used, then all wafers have equal chance of being measured, but out-of-specification lots are missed and yield decreases
Solution Approach 1:
The system performs preliminary actions by selecting specific wafers for metrology based on inspection results before the metrology measurement occurs. This allows the system to proactively identify and measure wafers that are likely to be out of specification, rather than relying on random selection. The inspection data is used in advance to guide metrology sampling decisions, ensuring that critical wafers are measured with higher probability.
Solution Approach 2:
The system implements feedback by using inspection results to inform and adjust metrology sampling decisions. The inspection data provides feedback about wafer quality, which is then used to modify the metrology sampling plan dynamically. This closed-loop approach ensures that metrology resources are allocated to wafers that most need measurement, improving detection reliability while maintaining productivity.
2Measurement precision
If dense sampling plan is used for metrology, then measurement accuracy improves, but throughput decreases and cost increases
Solution Approach 1:
The system applies local quality by using different sampling densities for different wafers based on their inspection results. Wafers identified as potentially out of specification receive denser sampling and measurement, while wafers appearing normal receive sparser sampling. This localized approach ensures measurement precision is applied where needed most, while maintaining throughput by reducing measurements on clearly acceptable wafers.
Solution Approach 2:
The system changes the sampling parameter (sampling density) dynamically based on inspection results. The sampling plan is adjusted from a fixed dense approach to a variable approach where the number of measurement points per wafer is modified according to the wafer's risk level. This parameter change allows the system to maintain high measurement precision for critical wafers while improving overall throughput by reducing measurements on non-critical wafers.
3Reliability
If metrology is performed at all wafer locations, then complete process monitoring is achieved, but time and cost increase significantly
Solution Approach 1:
The system extracts and focuses metrology measurements only on the critical subset of wafers identified through inspection, rather than measuring all wafers uniformly. By taking out the inspection function and using it to guide metrology sampling, the system achieves reliable process monitoring coverage for out-of-specification detection while significantly reducing the time and cost associated with measuring every single wafer.
Solution Approach 2:
The system applies partial action by performing metrology measurements on only a portion of the wafers - specifically those identified as high-risk through inspection. Rather than excessive action (measuring all wafers), the system uses partial sampling focused on critical areas, achieving sufficient process monitoring coverage for detecting out-of-specification lots while minimizing time loss.
Data Source
AI summary
Methods and systems for determining information for a specimen are provided. One method includes generating a sampling plan for only out of specification detection of a characteristic of specimens in a metrology process. The method also includes generating output for the specimens by performing the metrology process on the specimens with the generated sampling plan. In addition, the method includes determining the characteristic of the specimen based on the generated output and detecting if the characteristic of one or more of the specimens is out of specification based on the determined characteristic of the specimens. The embodiments described herein are particularly suitable for overlay metrology with substantially sparse sampling plans configured for only out of specification detection of the overlay.