Statistical Process Control for Semiconductor Uniformity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor manufacturing processes face challenges in achieving run-to-run uniformity and within-run uniformity due to variations in processing operations, which are critical as feature sizes decrease and complexity increases, leading to inconsistencies in semiconductor device formation.
Innovation Solution
Implementing statistical process control methods that involve collecting data from multiple sites on a substrate, analyzing it to identify outlier sites, and using algorithms to generate a comparative index to monitor and control processing operations, allowing for corrective actions to be taken based on trends and changes in the comparative index.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If statistical process control methods are implemented to monitor manufacturing equipment, then manufacturing precision and reliability are improved, but device complexity and measurement requirements increase
Solution Approach 1:
The patent introduces a comparative index as an intermediary metric that simplifies the monitoring process. Instead of directly analyzing complex multi-site measurement data, the system calculates a single comparative index value that captures the essential process state, making the monitoring system more manageable while maintaining precision
Solution Approach 2:
The patent transforms complex process data into a simplified parameter (comparative index) that can be easily monitored and controlled. This parameter transformation allows the system to track process uniformity without requiring complex analysis of all individual measurement points
2Manufacturing precision
If data is collected from multiple sites on substrates to ensure uniformity, then manufacturing precision is improved, but loss of time and measurement complexity increase
Solution Approach 1:
The patent extracts only the essential information needed for process control by identifying and monitoring the comparative index, rather than analyzing all raw data from multiple sites. This extraction approach maintains precision while reducing the time and effort required for data collection and analysis
3Manufacturing precision
If outlier sites are identified and monitored to control processing operations, then manufacturing precision is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The patent replaces complex manual outlier detection and analysis with an automated algorithmic system that calculates the comparative index. This substitution reduces the difficulty of detecting and measuring process variations by using systematic computational methods rather than manual inspection
Data Source
AI summary
A statistical process control method for monitoring and controlling semiconductor manufacturing processing operations is provided. For a chosen processing operation, multiple measurement sites are used to generate data of a measurable characteristic that is impacted by and associated with the processing operation. The data from the sites is compared over time and one or more outlier sites are identified. The outlier sites are the sites at which the data values are most divergent from the rest of the data. Algorithms are used to mathematically compare the outlier sites to the other sites to produce a comparative index. The comparative index is monitored graphically or otherwise to identify changes in the processing operation, and corrective actions are taken.