Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Apr 25;12(1):6711.
doi: 10.1038/s41598-022-08179-z.

Artificial intelligence deep learning for 3D IC reliability prediction

Po-Ning Hsu #  1   2 , Kai-Cheng Shie #  1   2 , Kuan-Peng Chen  3 , Jing-Chen Tu  4 , Cheng-Che Wu  1   2 , Nien-Ti Tsou  5 , Yu-Chieh Lo  1   2 , Nan-Yow Chen  6 , Yong-Fen Hsieh  7 , Mia Wu  7 , Chih Chen  1   2 , King-Ning Tu  8
Affiliations

Artificial intelligence deep learning for 3D IC reliability prediction

Po-Ning Hsu et al. Sci Rep. .

Abstract

Three-dimensional integrated circuit (3D IC) technologies have been receiving much attention recently due to the near-ending of Moore's law of minimization in 2D IC. However, the reliability of 3D IC, which is greatly influenced by voids and failure in interconnects during the fabrication processes, typically requires slow testing and relies on human's judgement. Thus, the growing demand for 3D IC has generated considerable attention on the importance of reliability analysis and failure prediction. This research conducts 3D X-ray tomographic images combining with AI deep learning based on a convolutional neural network (CNN) for non-destructive analysis of solder interconnects. By training the AI machine using a reliable database of collected images, the AI can quickly detect and predict the interconnect operational faults of solder joints with an accuracy of up to 89.9% based on non-destructive 3D X-ray tomographic images. The important features which determine the "Good" or "Failure" condition for a reflowed microbump, such as area loss percentage at the middle cross-section, are also revealed.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
(A) The overall circuit diagram of a test sample by using SAT. Where the rectangular area marked by yellow dashed lines is ROI. (B) Zoom-in image of one-fourth of ROI containing 100 microbumps. (C) CT image of the cross-section of a microbump.
Figure 2
Figure 2
SEM images and CT images in the XY, XZ, and YZ planes of a selected microbump at (A) the initial state, the state of the resistance increased by (B) 10% and (C) 20%.
Figure 3
Figure 3
The training/testing procedures and the performance of the CNN model.
Figure 4
Figure 4
The distributions of volume and area of the cross-section of all the 400 microbumps at their (A), (B) initial and (C), (D) reflowed states. (E) The distribution of the cross-section area loss percentage. (F) The volume loss of the entire microbump, (G) The area loss, and (H) The area loss percentage at the middle cross-section versus the initial volume of all the 400 microbumps.
See this image and copyright information in PMC

References

    1. Tu KN. Reliability challenges in 3D IC packaging technology. Microelectron. Reliab. 2011;51:517–523. doi: 10.1016/j.microrel.2010.09.031. - DOI
    1. Hsiao H-Y, et al. Unidirectional growth of microbumps on (111)-oriented and nanotwinned copper. Science. 2012;336:1007. doi: 10.1126/science.1216511. - DOI - PubMed
    1. Tu KN, Gusak AM. A unified model of mean-time-to-failure for electromigration, thermomigration, and stress-migration based on entropy production. J. Appl. Phys. 2019;126:075109. doi: 10.1063/1.5111159. - DOI
    1. Martins, A. et al. In 2018 International Wafer Level Packaging Conference (IWLPC) 1–6.
    1. Shulaker MM, et al. Three-dimensional integration of nanotechnologies for computing and data storage on a single chip. Nature. 2017;547:74–78. doi: 10.1038/nature22994. - DOI - PubMed

Publication types