Die Backside Optical Inspection for Bonding Defect Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in semiconductor manufacturing is the detection of defects such as tape residue and particle contaminants on semiconductor dies before bonding, which can lead to poor bonding quality and increased labor costs, breakage, and reduced yield.
Innovation Solution
An optical inspection system using a camera device and processor to inspect the backside surface of semiconductor dies for defects by capturing images of detecting spots and analyzing light reflection patterns to distinguish between defective and good dies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual inspection methods are used to detect defects on die backside surfaces, then labor flexibility is maintained, but inspection efficiency is low and labor costs are high
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated optical inspection system that uses light sources, cameras, and image processing algorithms to detect defects on die backside surfaces, thereby eliminating labor-intensive processes while maintaining inspection capability
Solution Approach 2:
The inspection system is integrated into the existing die handling process, where the die is automatically positioned and inspected without requiring separate manual handling steps, allowing the system to perform inspection as part of the normal manufacturing flow
2Reliability
If defects are not detected before bonding, then production continuity is maintained, but bonding quality deteriorates and yield decreases
Solution Approach 1:
The inspection system performs defect detection on the die backside surface before the bonding process occurs, allowing defective dies to be identified and removed in advance, preventing poor bonding quality and yield loss while maintaining production continuity
Solution Approach 2:
The inspection system focuses specifically on detecting defects on the backside surface of the die, which is the critical area that affects bonding quality, rather than inspecting the entire die structure, thereby optimizing detection effectiveness for the specific quality concern
3Manufacturing precision
If comprehensive defect inspection is performed on all dies, then quality is improved, but inspection time increases and productivity decreases
Solution Approach 1:
The inspection system extracts and inspects only the critical backside surface area of the die that directly impacts bonding quality, rather than performing comprehensive inspection of the entire die, thereby maintaining quality focus while reducing inspection time
Solution Approach 2:
The system performs inspection on the die backside surface with sufficient detail to detect bonding-critical defects, applying just enough inspection coverage to ensure quality without the time cost of exhaustive full-die inspection
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 system minimizes defects and contamination, reduces labor costs, and improves yield by identifying and excluding defective dies from the bonding process, thereby enhancing the quality and efficiency of semiconductor packaging.
Implementation Method 1
The camera device includes a light source for emitting light toward the die, and an optical sensor. The optical sensor is configured to detect an intensity of the light reflected by the one or more detecting spots
Data Source
AI summary
An optical inspection system includes a die carrier, a camera device, and a processor. The die carrier is configured to carry a die, wherein a backside surface of the die faces away from the die carrier. The camera device is movably disposed above the die carrier and configured to capture images of a plurality of detecting spots on the backside surface, wherein the plurality of detecting spots comprising a plurality of die corners of the die. The processor is coupled to the camera device and configured to determine a die center of the die according to the images of the plurality of die corners of the die, and determine whether the die is a bad die according to an existence of a defect on the backside surface of the die detected by the camera device.