Automatic detection device and method for semiconductor device and application

Abstract

The invention discloses an automatic detection device and method for a semiconductor device and application, the automatic detection device comprises a detection table, a first microscope, a second microscope and a control assembly, the first microscope and the second microscope are arranged vertically, and the detection table is used for bearing the semiconductor device at a view field intersection point of the first microscope and the second microscope. The detection table comprises a semiconductor adsorption assembly and a semiconductor moving assembly, and the control assembly is used for controlling the semiconductor moving assembly to translate and rotate the semiconductor adsorption assembly, controlling the first microscope and the second microscope to collect surface images of the semiconductor device and judging and recording detection data. The second microscope is a differential interference microscope. Polarized light is emitted to the side face of a semiconductor device to be detected through the second microscope and then reflected to an CCD imaging system of the second microscope to obtain a digital image, side face defects of the semiconductor device are extracted through an algorithm, and the high-precision detection of the semiconductor device is achieved.

Description

technical field [0001] The invention relates to the field of automatic detection of semiconductor devices, in particular to an automatic detection device, method and application of semiconductor devices. Background technique [0002] The manufacture of semiconductor devices usually involves doping, deposition, etching and other operations on a whole wafer. After the preparation of various transistors, dicing is performed, and the whole wafer is cut into individual semiconductor devices before packaging. At present, the inspection after dicing of semiconductor devices is a key node in quality control. At present, some manufacturers adopt the detection method of machine vision, which has the advantages of good detection automation and high efficiency. With the continuous improvement of semiconductor device manufacturing technology, the requirements for detection after scribing are getting higher and higher. Texture. The cleavage pattern is usually caused by the blunt cutting...

AI Technical Summary

Problems solved by technology

Existing conventional machine vision inspections mostly focus on taking pictures, and then use an algorithm to screen whether the smoothness of the inspection surface meets the standard, mainly including whether there are obvious impurities or cracks, and the cleavage lines on the inspection surface cannot be identified by direct focusing photography

Images