Substitute Sample Modeling for Semiconductor Process Parameter Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor processing, optimizing control parameters is costly and time-consuming due to the need for extensive learning data and the expense and time required for processing tests using actual product samples, which are difficult to measure efficiently.
Innovation Solution
A computer-based system that uses a substitute sample to correlate processing outputs with control parameters, allowing for the calculation of target control parameters for actual samples based on a model indicating the correlation between substitute and actual sample outputs, reducing the need for extensive learning data and costly actual sample measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large amount of learning data is used to optimize processing conditions through machine learning, then the accuracy of control parameter optimization is improved, but the process development cost and time increase due to the need for extensive processing tests using expensive actual product samples
Solution Approach 1:
The patent creates a substitute sample that copies the essential structural characteristics of the actual product sample but uses inexpensive materials. This substitute sample serves as a replica for measurement purposes, allowing the acquisition of learning data without using expensive actual product samples, thereby reducing both cost and time while maintaining measurement accuracy
Solution Approach 2:
The patent introduces a substitute sample made from inexpensive materials that can be used for multiple measurements without degrading. This substitute sample replaces expensive actual product samples in the measurement process, enabling the collection of large amounts of learning data at low cost and allowing rapid iteration for optimization
2Reliability
If actual product samples are used for processing tests to acquire learning data, then the reliability of optimization results is improved, but the measurement efficiency decreases due to the complexity and cost of actual samples
Solution Approach 1:
The substitute sample is designed to copy the critical structural features and measurement characteristics of the actual product sample while using simple, inexpensive materials. This allows measurements to be performed rapidly and efficiently on the substitute sample, which then translates to reliable optimization results when applied to the actual product
Solution Approach 2:
The patent changes the material parameters of the sample from expensive actual product materials to inexpensive substitute materials, while maintaining the structural and geometric parameters that are relevant to the measurement. This parameter substitution enables efficient measurement without sacrificing the reliability of the optimization results
Data Source
AI summary
A computer for determining a control parameter of processing to be performed on a sample includes: a memory unit configured to store a first model indicating a correlation between a first processing output obtained by measuring a first sample used for manufacturing, on which the processing is performed and a second processing output obtained by measuring a second sample that is easier to measure than the first sample and on which the processing is performed, and a second model indicating a correlation between a control parameter of the processing performed on the second sample and the second processing output; and an analysis unit configured to calculate a target control parameter of the processing performed on the first sample based on a target processing output as the target first processing output, the first model, and the second model.