3D Photoluminescence Model for Wafer Defect Origin Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for inspecting semiconductor and solar wafers fail to provide useful defect information to producers, limiting the ability to improve the quality of the bricks and wafers.
Innovation Solution
A method involving the creation of a three-dimensional photoluminescence model by combining sequential two-dimensional images of wafers, taken after slicing, to highlight defects and their origins within the brick, allowing for improved quality control and defect identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If 2D PL images of individual wafers are inspected separately, then defect detection is simple and fast, but useful defect information about the brick cannot be provided to producers
Solution Approach 1:
The patent combines multiple 2D PL images of individual wafers into a single 3D PL model representing the entire brick. This merging process integrates defect information from all wafers while preserving their spatial relationships, enabling producers to identify defect patterns and origins within the brick structure without inspecting each wafer separately.
Solution Approach 2:
The patent transitions from two-dimensional individual wafer inspection to three-dimensional brick-level analysis. By stacking 2D images in sequential order and assigning z-coordinates based on wafer position, the system creates a 3D model that adds the dimension of depth and spatial context, allowing producers to trace defects back to their origin locations within the brick.
2Loss of information
If 3D PL model construction is implemented to provide detailed defect information, then useful information for quality improvement is achieved, but processing time and computational complexity increase
Solution Approach 1:
The patent performs image stacking and 3D model construction as a preliminary processing step immediately after acquiring individual wafer images. By organizing images into sequential order and assigning spatial coordinates before detailed analysis, the system prepares the data structure in advance, reducing computational overhead during subsequent defect analysis and enabling faster retrieval of defect information.
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
The 3D model provides detailed information on defect locations and distributions within the brick, enabling producers to enhance the quality of both bricks and wafers by adjusting production processes or correcting defects in subsequent bricks.
Implementation Method 1
a two-dimensional (2D) photoluminescence (PL) image of each wafer is taken
Data Source
AI summary
Embodiments of the present invention generally relate to methods for inspecting wafers. After a brick is sliced into a plurality of bare wafers, a two-dimensional (2D) photoluminescence (PL) image of each wafer is taken, the PL images of the wafers in sequential order (i.e., the sequence of the wafers as they are sliced from the brick) are then combined to construct a three-dimensional (3D) model of the brick that highlights similar regions in the brick.