Silicon Carbide Wafer Defect Evaluation via KOH Etch Pit Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for evaluating crystal defects in silicon carbide single crystal wafers are limited by local observation, making it difficult to quantify the position, distribution, and degree of defect dense areas across the entire wafer surface.
Innovation Solution
A method involving etching the silicon carbide wafer with melted KOH to create etch pits, followed by automatic photography and image analysis to classify images as having or not having defect dense parts, allowing for precise evaluation of defect distribution and density.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If conventional microscope observation is used to evaluate crystal defects, then local observation capability is maintained, but the ability to evaluate the entire wafer surface and quantify defect distribution is lost
Solution Approach 1:
The wafer surface is divided into multiple observation regions, and microscopic images are automatically captured at multiple positions across the entire wafer surface. This segmentation approach enables comprehensive coverage of the large area while maintaining detailed microscopic observation capability, resolving the contradiction between observable area and measurement precision.
Solution Approach 2:
Multiple copies of microscopic images are automatically captured at different positions on the wafer surface. These image copies are then processed to evaluate defect distribution across the entire wafer, enabling both comprehensive area coverage and precise defect quantification simultaneously.
2Area of stationary object
If automatic photographing at multiple positions is implemented, then entire wafer surface coverage is achieved, but evaluation complexity increases
Solution Approach 1:
The evaluation system performs automatic image capture, processing, and defect evaluation without requiring manual intervention. The system automatically processes multiple images, determines presence of defect dense parts, and classifies images, thereby reducing evaluation complexity despite comprehensive wafer surface coverage.
Solution Approach 2:
Manual microscope observation and evaluation is replaced with an automatic photographing and image processing system. This substitution automates the entire evaluation process, maintaining simplicity while achieving complete wafer surface coverage through multiple position imaging.
3Measurement precision
If etch pit size is increased to 10-50 μm for better defect visibility, then defect detection precision is improved, but observation field of view is reduced
Solution Approach 1:
The wafer surface is divided into multiple observation regions, allowing large etch pits (10-50 μm) to be used in each region while still covering the entire wafer surface through multiple images. This segmentation resolves the contradiction between large field of view and high detection precision.
Solution Approach 2:
The evaluation transitions from a single two-dimensional field of view to a three-dimensional evaluation space by capturing images at multiple positions and processing them collectively. This dimensional expansion allows large etch pit sizes to be used while maintaining comprehensive coverage through multi-position imaging.
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 method enables high-precision evaluation of crystal defect density across the entire wafer surface, facilitating improved wafer quality and increased device yield by identifying and addressing defect dense areas.
Implementation Method 1
etching a silicon carbide single crystal wafer with melted KOH so that a size of an etch pit due to a threading edge dislocation is 10 to 50 μm
Data Source
AI summary
A method for evaluating crystal defects in a silicon carbide single crystal wafer, the method including steps of: etching a silicon carbide single crystal wafer with melted KOH so that a size of an etch pit due to a threading edge dislocation is 10 to 50 μm; obtaining microscopic images by automatic photographing at a plurality of positions on a surface of the silicon carbide single crystal wafer after the etching; determining presence or absence of a defect dense part in each of all the obtained microscopic images based on a continued length of the etch pit formed by the etching; and classifying all the obtained microscopic images into microscopic images having the defect dense part and microscopic images not having the defect dense part to evaluate a dense state of crystal defects in the silicon carbide single crystal wafer.