Substrate Lift Abnormality Detection From Electrostatic Chuck Load Signals

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional abnormality detection systems fail to quickly and accurately detect lifting abnormalities of substrates due to detachment failure from electrostatic chucks during substrate processing, leading to potential misalignment and damage.

Innovation Solution

An abnormality detection apparatus utilizing machine learning to analyze time-series measurements of load parameters applied to the lifting mechanism, generating a learning model that outputs a level of lifting abnormality, allowing for timely detection and prevention of transfer failures and substrate damage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional abnormality detection systems are used, then device complexity is reduced, but detection precision and speed of lifting abnormality are insufficient

Engineering Contradiction:
Improvedetection precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces conventional mechanical threshold-based detection systems with a machine learning-based detection system. The learning model analyzes time-series load parameter data to detect lifting abnormalities, achieving higher detection precision while maintaining acceptable system complexity through software-based intelligence rather than complex hardware arrangements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the detection approach from static threshold comparison to dynamic parameter analysis using machine learning. The system continuously monitors load parameters over time and uses a learning model to identify abnormal patterns, enabling faster and more accurate detection of lifting abnormalities compared to conventional methods.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If machine learning analysis is implemented, then detection speed and accuracy improve, but device complexity increases

Engineering Contradiction:
Improvedetection speedVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by pre-training the machine learning model with historical data before actual operation. The learning model is generated in advance through training with labeled data, enabling rapid real-time detection during actual substrate lifting operations without requiring complex real-time computation resources.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by training the learning model with replicated historical data patterns. The model learns from multiple examples of normal and abnormal lifting scenarios, creating a generalized detection capability that can quickly identify abnormalities in real-time without requiring the full complexity of analyzing every possible scenario from scratch.

Inventive Principle:
Principle #26Copying

3Reliability

If conventional detection methods are used, then system simplicity is maintained, but substrate damage prevention capability is insufficient

Engineering Contradiction:
Improvesubstrate protectionVSAvoidabnormality detection difficulty
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent implements feedback by continuously monitoring load parameters during substrate lifting and comparing real-time data against the trained learning model. The system provides immediate feedback when abnormalities are detected, enabling timely intervention to prevent substrate damage while maintaining operational awareness of lifting conditions.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent detects lifting abnormalities by analyzing vibrations and oscillations in the load parameters during substrate lifting. The machine learning model identifies characteristic vibration patterns that indicate detachment failure or abnormal lifting conditions, enabling detection of subtle abnormalities that conventional methods would miss.

Inventive Principle:
Principle #18Mechanical vibration

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

Enables rapid and accurate detection of lifting abnormalities, preventing substrate misalignment and damage by using machine learning to analyze load parameters, thereby improving the reliability of substrate processing systems.

Implementation Method 1

The substrate is electrostatically attracted to the electrostatic chuck by electrostatic force between the electrode and the substrate generated by applying a DC voltage to the electrode of the electrostatic chuck.

Methodology Applied
Scientific EffectElectrostatic force: Electrostatics

Data Source

PatentEP3796369B1Method for detecting abnormal lifting and lowering of substrate
Publication Date: 2024.03.27 SPP TECHNOLOGIES CO LTD
  • EP3796369B1 patent drawingFigure 1A~1B
  • EP3796369B1 patent drawingFigure 2
  • EP3796369B1 patent drawingFigure 3A~3C

AI summary

Provided is an abnormality detection apparatus and the like capable of quickly and accurately detecting lifting abnormality of a substrate attributable to detachment failure of a substrate from an electrostatic chuck, or the like. An abnormality detection apparatus 100 includes a measurement section 2 configured to measure a parameter having a correlation with load applied to a lifting mechanism 4; and a detection section 3 configured to detect lifting abnormality of a substrate S. The detection section 3 includes a learning model 31 generated by using machine learning, in which the learning model 31 receives, as input, a plurality of measurements of the parameter continuously measured by the measurement section 2 during lifting-up of the substrate S by the lifting mechanism 4, and outputs a level of lifting abnormality of the substrate S.