DocumentCode :
2811449
Title :
Impact of Radial defect clustering on 3D stacked IC yield from wafer to wafer stacking
Author :
Singh, Eashendra
fYear :
2012
fDate :
5-8 Nov. 2012
Firstpage :
1
Lastpage :
7
Abstract :
We present and evaluate a simulation methodology to model the impact of the radial clustering of defects on wafers on the yield of 3D ICs manufactured using wafer to wafer stacking. Our simulations draw on an extensively validated model for radial yield degradation on wafers from the literature to incorporate the effect of this key contributor to the widely observed clustering of defects on semiconductor wafers. Current 3D-SIC yield estimation methods ignore defect clustering and assume a uniform distribution of defective dies on each wafer. Our results show that the radial clustering of defective dies causes stacked die yields to be significantly higher than that projected by current models. For 4-5 layer stacks the difference can be 50% or more. Our simulation studies are further validated by comparison with actual silicon data, which suggests that even the more accurate yield estimates from our improved methodology may be somewhat pessimistic. Thus the results presented here show that in practice degradation in stacked die yield from compounding may not be as severe as commonly estimated. This has significant implications in evaluating cost trade-offs associated with 3D-SIC manufacturing.
Keywords :
integrated circuit packaging; integrated circuit yield; 3D stacked IC yield; 3D-SIC manufacturing; 3D-SIC yield estimation methods; defective dies; radial clustering; radial defect clustering; radial yield degradation; semiconductor wafers; silicon data; simulation methodology; stacked die yields; wafer to wafer stacking; wafers defects; Degradation; Fabrication; Integrated circuit modeling; Mathematical model; Predictive models; Semiconductor device modeling; Stacking;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Test Conference (ITC), 2012 IEEE International
Conference_Location :
Anaheim, CA
ISSN :
1089-3539
Print_ISBN :
978-1-4673-1594-4
Type :
conf
DOI :
10.1109/TEST.2012.6401567
Filename :
6401567
Link To Document :
بازگشت