2013 IEEE 15th Electronics Packaging Technology Conference (EPTC 2013) 2013
DOI: 10.1109/eptc.2013.6745735
|View full text |Cite
|
Sign up to set email alerts
|

Microstructure investigation of Cu/SnAg solid-liquid interdiffusion interconnects by Electron Backscatter Diffraction

Abstract: The investigation of the grain structure of Cu/SnAg interconnects produced by solid-liquid interdiffusion (SLID) bonding is presented in this study. The texture analysis was carried out using Electron Backscatter Diffraction (EBSD). The samples were manufactured by flux-assisted bonding of two Si dies with an area array of square Cu/SnAg bumps on the bottom die and square Cu bumps on the top die at a temperature of 250°C and bonding time between 35 s and 40 s. The influence of the bonding process carried out u… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

1
3
0

Year Published

2015
2015
2024
2024

Publication Types

Select...
5
1

Relationship

1
5

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 8 publications
1
3
0
Order By: Relevance
“…Note that for microelectronic packaging applications, Cu and Sn layers are deposited by electrochemical process. In these cases, in addition to the classical Kirkendall voids defects, numerous porosities and large cavities are observed at end of TLPB process inside the joints as it was reported by several studies [1,[22][23][24]. At this stage, it is important to note that these large cavities observed in cross-section analyses at the end of TLPB were not observed at the initial stage of the process.…”
Section: Introductionsupporting
confidence: 65%
“…Note that for microelectronic packaging applications, Cu and Sn layers are deposited by electrochemical process. In these cases, in addition to the classical Kirkendall voids defects, numerous porosities and large cavities are observed at end of TLPB process inside the joints as it was reported by several studies [1,[22][23][24]. At this stage, it is important to note that these large cavities observed in cross-section analyses at the end of TLPB were not observed at the initial stage of the process.…”
Section: Introductionsupporting
confidence: 65%
“…The “no contact” area is detected in the upper right die corner for most of the bonded sample pairs (the orientation of the dies was marked). This suggests that there may be a slight non-planarity between the bonding stage and head, which is a known issue for bonding in the literature [ 41 , 42 , 43 ]. This could lead to variations in fracture surface types and bonding strength across the sample pairs.…”
Section: Resultsmentioning
confidence: 99%
“…Shrinkage voids have also been discussed for the formation of Cu 6 Sn 5 and Cu 3 Sn IMCs in the Cu=Sn SLID system associated with a volume shrinkage of 5.23 and 4.39% respectively. 6) Since the density of liquid In and Sn at bonding temperature is much lower (7.03 g•cm −3 for In at 164 °C29) and 6.98 g•cm −3 for Sn at 232 °C31) ) compared to RT, the resulting volume shrinkage increases to 10.68% for Cu 11 In 9 and 7.83% for Cu 6 Sn 5 IMCs.…”
Section: Cu/in Bonding and High Temperature Storagementioning
confidence: 99%
“…The resulting bonding technologies are either thermo-compression bonding (TCB, for Cu=Cu) 4,5) or solid-liquid interdiffusion bonding (SLID, for Cu=Sn=Cu). 6,7) SLID can be easily implemented since it is a robust process and it does not require very high bonding pressures or temperatures compared to Cu=Cu TCB. A thin layer of a low-melting metal is typically electroplated on the top of a high-melting metal.…”
Section: Introductionmentioning
confidence: 99%