Method for analyzing BEOL testing chip on-line failure

Abstract

The invention discloses a method for online failure analysis of a BEOL test chip. The invention generates a plurality of virtual defects by splitting the failure structure of the characteristic test, and synthesizes the detection result of the defect, so as to realize the search, confirmation and analysis of the failure position by the automatic automatic scanning electron microscope. The invention can avoid the destructive analysis of the silicon chip, improve the analysis efficiency and success rate, realize the real-time feedback to the detection program at the same time, and improve the efficiency and accuracy of the program optimization. The invention is suitable for the online failure analysis method of the BEOL test chip in the semiconductor process.

Description

technical field [0001] The invention relates to a semiconductor manufacturing process, in particular to a method for online failure analysis of a BEOL test chip in the semiconductor process. Background technique [0002] The failure analysis of the test chip plays a very important role in the semiconductor process, and is generally performed after the last layer of metal lines is completed. The methods used are mainly De-Layer by chemicals, Polishing by chemical mechanical grinding, and destructive means such as Focused Ion Beam (FIB). Moreover, in order to ensure a certain capture rate for commonly used test structures such as open and short (Open&Short), the size of the test structure is generally more than 1000 microns. It will be very time-consuming and laborious to search manually and the success rate is very low. [0003] With the failure analysis method of the test chip in the above prior art, although its final test result can reflect the capture rate of each detect...

AI Technical Summary

Problems solved by technology

Moreover, in order to ensure a certain capture rate for commonly used test structures such as open and short (Open&Short), the size of the test structure is generally above 1000 microns. It will be very time-consuming and laborious to search manually and the success rate is very low.

[0003] With the failure analysis method of the test chip in the above prior art, although its final test result can reflect the capture rate of each detection layer, it cannot pass point-to-point for the undetected defects found in failure analysis (failure analysis, hereinafter referred to as FA). program adjustments to optimize

[0004] The existing back end of line (BEOL) test chip online failure analysis method of the existing chip production line has the characteristics of strong destructiveness, low efficiency, and low success rate, and it cannot target the unidentified problems discovered by FA. Detected defects are adjusted and optimized, and there are shortcomings such as insufficient feedback for detection program optimization

Images