AI Wafer Rotation Using In-Line Metrology for Yield Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional semiconductor manufacturing processes face challenges in defect root cause analysis due to variability across tool manufacturers, requiring extensive cross-disciplinary coordination and prolonged diagnostic timelines, leading to production loss and inefficiencies.

Innovation Solution

An artificial intelligence-based wafer rotation method and device that integrates data measurement, rotation angle calculation, and wafer alignment to optimize manufacturing yield by dynamically adjusting wafer orientation at each process step, using integrated metrology and machine learning algorithms to identify and mitigate defects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional wafer rotation methodologies are used (RCP-based or equipment shutdown methods), then defect root cause analysis can be performed, but production downtime increases and manufacturing yield decreases

Engineering Contradiction:

Improvedefect root cause analysis capabilityVSAvoidmanufacturing yield

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system performs wafer rotation and measurement operations in advance during manufacturing processes, so that defect analysis data is already available when needed, eliminating the need for production shutdowns and enabling continuous manufacturing while maintaining defect analysis capability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

An AI-based intermediary system processes measurement data from manufacturing equipment to identify defect patterns and root causes, replacing the need for direct equipment shutdown and manual cross-disciplinary analysis, thereby maintaining production continuity while enabling effective defect analysis

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If conventional wafer rotation methodologies are used, then defect analysis can be conducted, but diagnostic timeline is prolonged

Engineering Contradiction:

Improvedefect analysis capabilityVSAvoiddiagnostic timeline

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual, mechanical defect analysis procedures with an AI-based automated system that processes measurement data to rapidly identify defect patterns and root causes, dramatically reducing diagnostic time while maintaining or improving analysis accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service defect analysis by automatically processing measurement data and generating defect root cause identification without requiring extensive cross-disciplinary coordination or manual intervention, thereby accelerating the diagnostic process

Inventive Principle:

Principle #25Self-service

3Productivity

If AI-based wafer rotation is implemented, then manufacturing yield is optimized and defect propagation is minimized, but system complexity increases

Engineering Contradiction:

Improvemanufacturing yieldVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements a universal AI-based system that can perform multiple functions including wafer rotation, measurement data processing, defect pattern recognition, and root cause identification across different process steps, reducing the need for multiple separate systems and thereby managing complexity while enhancing yield optimization capabilities

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250336706A1Device and method with artificial intelligence-based wafer rotation

Publication Date: 2025.10.30 SAMSUNG ELECTRONICS CO LTD

AI summary

An artificial intelligence-based wafer rotation method according to an embodiment includes receiving measurement data of a manufacturing process from a data measurement module integrated into manufacturing equipment, calculating rotation angle of a wafer for each process step that maximizes manufacturing yield of the manufacturing process based on the measurement data, and rotating the wafer at the calculated rotation angle in at least one of process steps.