SiC Wafer Defect Mapping With Nondestructive Optical Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional defect characterization methods for silicon carbide (SiC) wafers, such as etching, are destructive, expensive, and limited in providing comprehensive defect information, hindering the development of high-quality, large-diameter SiC wafers for advanced semiconductor devices.
Innovation Solution
Nondestructive defect characterization using deep convolutional neural networks (DCNNs) for SiC wafers, enabling defect detection and identification based on nondestructive data, such as photoluminescence microscopy, and training the network with destructive data from etched wafers to predict defects without etching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If destructive etching is used for defect characterization, then defect detection accuracy is improved, but wafer loss and cost increase
Solution Approach 1:
The patent creates a digital copy (defect map) of the wafer surface using nondestructive optical imaging. This defect map contains all necessary defect information without requiring physical alteration of the wafer. The copying principle allows multiple analyses of the same wafer without loss, resolving the contradiction between accurate defect detection and wafer preservation.
Solution Approach 2:
The patent replaces the mechanical/chemical etching process with an optical imaging system. Instead of using chemicals to physically remove material and reveal defects, the system uses light to nondestructively image defects on the wafer surface. This substitution eliminates wafer loss while maintaining defect detection capability.
2Measurement precision
If destructive etching is used for defect characterization, then defect identification is improved, but process complexity and cost increase
Solution Approach 1:
The patent replaces the complex multi-step etching process with a single optical imaging step. The automated defect detection system uses software algorithms to identify and characterize defects directly from images, eliminating the need for chemical etching, rinsing, drying, and manual inspection steps. This reduces process complexity while maintaining or improving defect identification accuracy.
Solution Approach 2:
The system performs automated defect detection and characterization without requiring manual intervention. The software automatically analyzes the optical images, identifies defect types, and generates defect maps. This self-service capability simplifies the overall process by eliminating manual etching and inspection steps, reducing both process complexity and operational costs.
3Quantity of substance
If destructive etching is used for defect characterization, then sampling is performed, but information availability is limited
Solution Approach 1:
The patent creates comprehensive digital copies (defect maps) of entire wafer surfaces, capturing all defect information in a nondestructive manner. Unlike destructive sampling that examines only etched portions, the optical imaging system can capture the entire wafer surface, providing complete information availability while enabling sampling of more wafers since none are destroyed in the process.
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 high-fidelity defect characterization and counting of SiC wafers nondestructively, reducing costs, preserving wafers for device fabrication, and enhancing feedback loops for crystal growth processes.
Implementation Method 1
capturing the nondestructive data comprises photoluminescence microscopy
Data Source
AI summary
A method of analyzing semiconductor wafers includes capturing a first image of a first crystalline material, etching a first surface of the first crystalline material to delineate etch defects in the first crystalline material, and capturing a second image of first crystalline material after etching the first surface of the first crystalline material. Based on the second image, labels of etch defects delineated in the first surface of the first crystalline material are generated. The first image and the labels of etch defects are spatially coordinated to form a defect map identifying one or more defects in the first image based on the delineated etch defects, and based on the defect map and nondestructive data obtained from a second crystalline material, defects in the second crystalline material are identified.