Defect source analysis method and analysis system, defect detection device

Abstract

The present invention relates to a defect source analysis method, analysis system, and defect detection device. The process layer of the target defect is obtained by optically scanning the wafer and processed with several previously scanned process layers, and the first several layers are judged according to the sequence from near to far. Whether there is a characteristic defect matching the target defect in each scanned process layer, so as to analyze the source of the target defect to obtain the process layer (or corresponding process) that introduces the target defect, that is, to obtain the target defect The source of the defect source, through the analysis method, analysis system and defect detection device of the defect source provided by the present invention, the man-hours spent on manually judging the source of the bubble defect and the manual error caused are greatly reduced. Through the analysis and monitoring of the source of the bubble defect, The abnormal situation of the process or machine can be determined in time, which helps to reduce unnecessary losses.

Description

technical field [0001] The invention relates to the technical field of semiconductors, in particular to an analysis method, an analysis system and a defect detection device for defect sources. Background technique [0002] With the continuous maturity and development of semiconductor manufacturing technology, image sensors are increasingly used in many fields such as digital cameras, PC cameras, video phones, video conferencing, intelligent security systems, car reversing sensors, game consoles, and industrial medicine. [0003] Image sensors can be divided into CCD (Charge Coupled Device, Charge Coupled Device) image sensors and CMOS (Complementary Metal Oxide Semiconductor, complementary metal-oxide-semiconductor) image sensors according to the difference between photosensitive elements and photosensitive principles. Among them, the CMOS image sensor belongs to the optoelectronic component and the CMOS image sensor can integrate the driving circuit and the pixel because it...

AI Technical Summary

Problems solved by technology

[0005] However, after using the above-mentioned method to bond the bottom wafer and the device wafer, the combined wafer is optically inspected by an optical scanning machine, and there are bubble defects in the formed wafer.

However, the existing image analysis system has errors in the analysis of bubble defects, and it is difficult to determine the source of bubble defects

Bubble defects may be introduced by the underlying wafer, or by device wafers or other process layers (or corresponding processes). Currently, judging the source of bubble defects requires manual judgment and manual production of bubble defect trend charts. This process increases labor costs and has manual errors, and it is also difficult to detect and monitor abnormal conditions of the process or machine in time

Images