Reflectometry Endpoint Detection With Fault Classification for Plasma Etch
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current endpoint detection (EPD) systems for reflectometry in semiconductor fabrication are complex and lack fault detection and classification (FDC) capabilities, making troubleshooting difficult and accuracy dependent on wafer patterns, with no self-contained solutions available for plasma etch processes.
Innovation Solution
A method and apparatus for endpoint detection using a wafer classification model to determine wafer and product types, execute EPD algorithms, and perform fault diagnosis through etching processes, including post-etching outlier models to identify and correct system faults.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fault detection and classification capabilities are added to EPD systems, then system reliability is improved, but device complexity increases
Solution Approach 1:
The FDC system is segmented into multiple independent modules: incoming outlier model for pre-etch fault detection, etching rate model for process monitoring, and post-etch outlier model for final verification. Each module handles specific fault detection tasks independently, improving reliability without overwhelming system complexity.
Solution Approach 2:
The incoming outlier model performs preliminary fault detection before the etching process begins, identifying potential issues with wafers, optical instruments, or tooling early in the workflow. This preventive approach improves reliability by catching faults before they affect production, while keeping the main EPD system unchanged.
2Measurement precision
If multiple models and algorithms are integrated for comprehensive fault detection, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The detection system is divided into three specialized models: incoming outlier model for pre-process faults, etching rate model for process deviations, and post-etch outlier model for final product verification. Each model focuses on specific measurement aspects, improving overall detection precision while managing complexity through functional separation.
Solution Approach 2:
A centralized controller acts as an intermediary that coordinates the multiple models and algorithms. It receives data from various sensors, routes information to appropriate models, and integrates their outputs into unified fault detection decisions, improving measurement precision without requiring direct complex interactions between all components.
3Productivity
If real-time monitoring and fault classification are implemented, then productivity is improved through reduced downtime, but device complexity increases
Solution Approach 1:
The system implements continuous feedback loops where sensor data from the etching process is constantly monitored by the etching rate model, which compares actual rates against expected values. When deviations are detected, the system automatically classifies the fault type and triggers appropriate responses, improving productivity through rapid fault resolution while using standardized feedback mechanisms.
Solution Approach 2:
The FDC system performs self-diagnosis and automatic fault classification without requiring external intervention. The incoming and post-etch outlier models autonomously evaluate wafer and product quality, while the etching rate model self-monitors process health, reducing the need for complex external control systems and improving productivity through automated decision-making.
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
Improves manufacturability, in-situ performance, and reliability of reflectometry EPD systems by identifying faults before they occur, reducing system downtime, and achieving best possible accuracy with minimal engineer input.
Implementation Method 1
Initial data of a wafer measured by an optical instrument are received
Implementation Method 2
an etching rate EPD algorithm is executed based on emission optical spectroscopy (OES) data and voltage-current (VI) data to obtain the etching rate
Data Source
AI summary
A method of endpoint detection includes receiving a wafer classification model that includes wafer types, product types and end-point detection (EPD) algorithms. Each EPD algorithm corresponds to a respective wafer type and a respective product type. Initial data of a wafer measured by an optical instrument are received. The wafer classification model is executed based on the initial data to determine a wafer type, a product type and an EPD algorithm for the wafer or to determine whether the wafer or the optical instrument is faulty. An etching process is executed on the wafer to obtain a product. The EPD algorithm is run to obtain an etching depth using data associated with the etching process so that an endpoint time of the etching process is determined by the etching depth or a maximum endpoint time. A post-etching outlier model is executed to determine whether the product is faulty.