Plasma RF Frequency Inference for Low Reflective Wave Power
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing plasma processing technologies face challenges in minimizing the power of reflective waves reflected from the plasma processing space, which affects the stability of plasma and efficiency of substrate processing.
Innovation Solution
A plasma processing apparatus that includes an inference circuitry to infer the optimal radio frequency frequency for plasma generation by inputting processing conditions into a learned model trained on data that minimizes reflective wave power, allowing for the adjustment of microwave and radio-frequency settings to stabilize plasma.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the frequency of bias radio frequency is changed to search for optimal frequency, then the power of reflective waves can be minimized, but the time required for frequency search and system complexity increases
Solution Approach 1:
The system performs preliminary frequency search and learns the relationship between processing conditions and optimal frequencies in advance, storing this knowledge in a database. During actual plasma processing, the pre-learned optimal frequency is directly retrieved based on the current processing conditions, eliminating the need for real-time frequency searching and significantly reducing the time required.
Solution Approach 2:
The patent replaces the mechanical trial-and-error frequency tuning process with an information processing system that uses learned models and databases to automatically determine optimal frequencies. The control device queries the database with current processing conditions and receives the optimal frequency setting, substituting physical frequency sweeping with computational lookup and inference.
2Reliability
If frequency search is performed for each processing condition, then optimal plasma stability can be achieved, but the complexity of control and measurement increases
Solution Approach 1:
The system performs the complex frequency search and analysis work in advance during a learning phase, building a database that maps processing conditions to optimal frequencies. During actual plasma processing, the control complexity is dramatically reduced as the system simply queries the pre-built database with current conditions and applies the retrieved optimal frequency, maintaining plasma stability without real-time searching.
Solution Approach 2:
The patent creates a virtual model of the frequency-search process by learning and storing the relationship between processing conditions and optimal frequencies in a database. This virtual model (copy of the optimal settings) is then used during actual processing, allowing the system to replicate the benefits of extensive frequency searching without performing the actual complex measurements and adjustments in real-time.
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 enables the efficient search for and implementation of a set frequency that minimizes reflective wave power, enhancing plasma stability and processing efficiency by fine-tuning the plasma generation settings based on learned models.
Implementation Method 1
a radio frequency for generating plasma is set
Data Source
AI summary
A plasma processing apparatus including: an inference part that infers a set frequency of a radio frequency for generating plasma by inputting a processing condition for performing a plasma processing into a learned model that has been trained using learning data including the set frequency that minimizes power of a reflective wave reflected from a processing space and a processing condition corresponding to the set frequency, the set frequency being searched by changing the set frequency of the radio frequency for generating plasma when a plasma processing is performed under each of a plurality of processing conditions.