AI Etching Recipe Feedback for Wafer Yield Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The scaling down of semiconductor devices poses challenges in achieving improved quality, yield, performance, and reliability while reducing complexity, particularly in the etching process.
Innovation Solution
A method for controlling an etching tool using an artificial intelligence module that executes etching recipes, generates and analyzes data from measurement modules, and adjusts recipes in real-time based on predetermined ranges to optimize the etching process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If semiconductor device dimensions are scaled down to improve computing ability, then device performance is improved, but manufacturing complexity and process control difficulty increase
Solution Approach 1:
The patent implements a feedback control system where process data from etching operations is collected, analyzed, and used to generate adjusted recipes. The system monitors etching results and feeds this information back to automatically modify subsequent etching parameters, thereby managing process control complexity while maintaining improved device performance through precise dimensional control.
Solution Approach 2:
The etching tool system performs self-optimization by automatically analyzing its own process data and generating adjusted recipes without external intervention. The artificial intelligence module enables the system to self-adjust etching parameters based on real-time and historical data, reducing the need for manual process tuning and simplifying control as dimensions are scaled down.
2Device complexity
If traditional etching control methods are used, then process simplicity is maintained, but manufacturing precision and yield deteriorate due to process variations
Solution Approach 1:
The system collects process data from etching operations and uses this feedback to automatically adjust etching recipes. By monitoring actual etching results and comparing them to target specifications, the system generates corrected recipes that compensate for process variations, thereby maintaining high manufacturing precision while keeping the control approach automated rather than manually complex.
Solution Approach 2:
The artificial intelligence module dynamically adjusts etching parameters such as power, pressure, and gas flow based on analyzed process data. This automatic parameter modification allows the system to maintain precise etching control despite variations in materials, equipment, or environmental conditions, improving manufacturing precision without requiring overly complex control procedures.
3Manufacturing precision
If real-time data analysis and recipe adjustment are implemented, then manufacturing precision and yield are improved, but device complexity and system complexity increase
Solution Approach 1:
The artificial intelligence module serves multiple functions: it analyzes process data, generates adjusted recipes, controls etching parameters, and manages the overall etching process. By consolidating these functions into a single multi-functional system, the patent improves manufacturing precision through real-time control while avoiding the need for multiple separate complex systems, thereby managing overall system complexity.
4Productivity
If artificial intelligence-based recipe adjustment is used, then productivity and yield are improved through reduced rework, but device complexity increases
Solution Approach 1:
The system uses feedback from process monitoring to automatically adjust recipes and prevent defects before they occur. This real-time correction capability reduces the need for rework and scrap, thereby improving productivity and yield. The automated feedback loop eliminates the need for manual intervention and reprocessing, offsetting the complexity of the AI system with gains in operational efficiency.
Solution Approach 2:
The artificial intelligence module performs preliminary analysis of process data and makes preventive adjustments to recipes before problems manifest. By anticipating and correcting potential issues in advance, the system prevents yield losses and reduces the need for corrective rework, thereby improving productivity while the automated nature of the preliminary actions manages the complexity burden.
Data Source
AI summary
The present application discloses a method for controlling an etching tool. The method includes executing a first etching recipe on a current wafer; generating a first set of data of the current wafer by a first measurement module; analyzing the first set of data by an artificial intelligence module coupled to the first measurement module; generating, by the artificial intelligence module, a second etching recipe and applying the second etching recipe to the etching tool when the first set of data is not within a predetermined range; and executing the second etching recipe on a next wafer.