Automated Plasma Characterization via Curve-Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for characterizing plasma during semiconductor substrate processing are time-consuming and inefficient, requiring several minutes to hours to analyze data from thousands or millions of RF bursts, making it difficult to provide timely relevant data for process control.
Innovation Solution
An automated method that identifies a relevancy range and determines seed values to perform curve-fitting, allowing for rapid characterization of plasma parameters within a few milliseconds, eliminating the need to analyze the entire data set and reducing manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the entire data set from thousands or millions of RF bursts is analyzed manually, then comprehensive plasma characterization is achieved, but the processing time increases to several minutes or hours
Solution Approach 1:
The patent segments the large data set from thousands or millions of RF bursts into smaller, manageable subsets. By dividing the comprehensive data into segments, the system can process each segment individually and efficiently, reducing the overall processing time from hours to minutes while maintaining characterization accuracy through systematic analysis of all segments.
Solution Approach 2:
The patent applies partial action by analyzing representative subsets of the data rather than the entire data set. By selecting and processing key portions of the data that contain sufficient information for plasma characterization, the system achieves effective plasma analysis without the time cost of processing every single data point.
2Productivity
If automated curve-fitting with relevancy range identification is implemented, then processing time is reduced to milliseconds, but system complexity increases
Solution Approach 1:
The patent implements self-service through automated algorithms that independently identify relevancy ranges and perform curve-fitting without manual intervention. The system automatically processes the data, fits curves to identify plasma parameters, and generates characterization results, eliminating the need for manual analysis while achieving millisecond processing times.
Solution Approach 2:
The patent changes the approach from analyzing all data points to identifying and analyzing only the relevancy range - a specific parameter subset of the data that contains the essential information for plasma characterization. This parameter change enables rapid processing while maintaining accuracy.
3Device complexity
If manual data analysis is performed, then system complexity is minimized, but the ability to provide timely process control data is reduced
Solution Approach 1:
The patent replaces the mechanical manual analysis process with an automated computational system. By substituting human manual operations with automated algorithms for relevancy range identification and curve-fitting, the system achieves rapid millisecond processing that provides timely feedback for process control while maintaining operational simplicity through automation.
Data Source
AI summary
A method for automatically characterizing plasma during substrate processing is provided. The method includes collecting a set of process data, which includes at least data about current and voltage. The method also includes identifying a relevancy range for the set of process data, wherein the relevancy range includes a subset of the set of process data. The method further includes determining a set of seed values. The method yet also includes employing the relevancy range and the set of seed values to perform curve-fitting, wherein the curve-fitting enables the plasma to be automatically characterized.