Wafer Micro-Crack Detection Using Angled Infrared Line-Scan Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional high-speed imaging systems are limited in detecting micro-cracks in solar wafers due to insufficient image resolution and the complexity of multi-crystalline wafer textures, which makes it difficult to identify cracks smaller than 36 μm and those with random surface features.

Innovation Solution

The method involves directing infrared light along two angled axes to capture high-contrast images of solar wafers, using a pair of line-scan imaging devices and light sources positioned at acute angles to enhance image resolution and detect micro-cracks of varying orientations, with image processing to merge and analyze the captured images for defect detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a conventional high-speed imaging system with a 12K line-scan camera is used to inspect solar wafers, then the inspection speed can maintain high productivity (one wafer per second), but the image resolution is insufficient to detect micro-cracks smaller than 36 μm

Engineering Contradiction:
Improveinspection speedVSAvoidmicro-crack detection capability
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent divides the inspection task into multiple sequential line-scan acquisitions across different regions of the wafer. Instead of attempting to capture the entire wafer in a single high-resolution image, the system segments the inspection into multiple passes, allowing each line-scan to focus on capturing fine details of micro-cracks while maintaining overall inspection coverage and productivity.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If the image resolution is increased to detect micro-cracks smaller than 36 μm, then the measurement precision improves, but the inspection speed decreases due to the complexity of processing higher resolution images

Engineering Contradiction:
Improvemicro-crack detection capabilityVSAvoidinspection speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent introduces a temporal dimension to the inspection process by performing multiple line-scan acquisitions over time. Instead of increasing spatial resolution in a single frame, the system accumulates information across multiple temporal samples, effectively achieving high precision detection while maintaining processing efficiency through sequential rather than simultaneous data collection.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Difficulty of detecting and measuring

If infrared light is used to penetrate the solar wafer for internal structure examination, then the detection capability for micro-cracks improves, but the image contrast is reduced due to the complex multi-crystalline wafer textures

Engineering Contradiction:
Improvemicro-crack detectabilityVSAvoidimage contrast
Core Design Contradiction:
Difficulty of detecting and measuringVSIllumination intensity

Solution Approach 1:

The patent extracts and isolates the micro-crack signals from the complex multi-crystalline background by using line-scan imaging to capture sequential horizontal or vertical lines across the wafer. This approach separates the defect detection task from the challenging texture analysis, allowing micro-cracks to be identified through their distinct linear patterns even against complex crystalline backgrounds.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the illumination and detection parameters by using infrared wavelengths that penetrate silicon, combined with line-scan acquisition modes that capture data along specific orientations. This parameter change allows the system to detect micro-cracks independent of their actual width and orientation, overcoming the contrast reduction caused by multi-crystalline textures.

Inventive Principle:
Principle #35Parameter changes

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 approach allows for the detection of micro-cracks as small as 1 μm, independent of their actual width, and provides high-contrast images that facilitate the identification of cracks extending randomly, improving the detection of micro-cracks in both mono- and multi-crystalline solar wafers.

Implementation Method 1

due to silicon's band-gap energy level, solar wafers appear transparent when illuminated with light having a wavelength larger than 1127 nm. Light having a wavelength of 1127 nm is classified as near infrared (NIR) radiation. NIR is invisible to the human eye but is detectable by most commercial CCD or CMOS infrared cameras.

Methodology Applied
Scientific EffectInfrared radiation penetration: Infrared Radiation

Data Source

PatentUS8428337B2Apparatus for detecting micro-cracks in wafers and method therefor
Publication Date: 2013.04.23 APPLIED MATERIALS INC
  • US8428337B2 patent drawing
  • US8428337B2 patent drawing
  • US8428337B2 patent drawing

AI summary

A method and apparatus for wafer inspection is disclosed. The method and apparatus involve directing light substantially along a first axis towards a first surface of a wafer to thereby obtain light emanating along the first axis from a second surface of the wafer, wherein the first and second surfaces of the wafer are substantially outwardly opposing and substantially extending parallel to a plane. The method and apparatus further involve directing light substantially along a second axis towards the first surface of the wafer to thereby obtain light emanating along the second axis from the second surface of the wafer, the first axis being angled away from the second axis about a reference axis extending along the plane. More specifically, the orthographic projection of the first axis on the plane is substantially parallel to the orthographic projection of the second axis on the plane, and each the orthographic projections of the first and second axes on the plane is substantially orthogonal to the reference axis.