Guest
 Posted: Mon Jan 26, 2026 10:55 am Post subject: Au-Si eutectic bonding |
 |
| I would appreciate it very much if somebody can give me a good insight into the mechanism of Au-Si eutectic bonding.
From what I understand during Au-Si eutectic bonding, the gold and silicon interface melts into each other at 363C to form a Au-Si eutectic. And if we do a die shear test and find good silicon remains on the paddle, we conclude that there is good eutectic formation. What I am curious to know is what happens at the interface between the gold (from the die back) and Ag plating on the paddle? In order to have good die shear strength, isn't the adhesion between the gold and silver just as important? I am assuming that in the Au-Si die attach mechanism, there are 2 interfaces involved---1 interface between the silicon & gold backing and another interface the gold backing & the Ag plating (from the lead). Is this thinking correct? Or is there only 1 interface after the melting? Would appreciate your help. Thank you and regards. |
|
Joined: 16 Jun 2026
Posts: 108
Location: Paranaque City, Phils.
| Posted: Wed Jan 28, 2026 12:45 am Post subject: |
|
| Hi Curious,
The Au-Si eutectic processes that I'm familiar with use a gold substrate inside the cavity, over which a gold preform is placed and pre-heated before the die is positioned over it. The entire system is then heated up to reach the eutectic temp, at which point the preform starts to melt because the Si atoms from the die backside will already have diffused into it. Scrubbing the die distributes the die attach evenly under the backside. Eventually the entire preform melts and becomes integrated into the die pad of the cavity and the die backside itself. As the Si concentration exceeds the eutectic composition, the melting temp of the die attach goes up again and the die attach material solidifies. What you get is a uniform eutectic die attach system - so uniform in fact that it would no longer be possible to find an 'interface' between the Si and Au. EE |
|
  |
Guest
| Posted: Wed Jan 28, 2026 12:51 pm Post subject: Au-Si eutectic bonding |
|
| Hello EE,
Thank you very much for your information. In our Au-Si bonding, we have 2 ways. If the gold backing on the wafer is not thick enough, we use gold preform like you mentioned. The die is then bonded on the tab with scrubbing and under high temperature. But if the gold backing on the wafer back is sufficiently thick, we do a direct bonding (no gold preform) and there is no srubbing here--just pick and place (under high temperature, of course) I understand that the silicon substrate and the gold will melt into each other to form a eutectic. What I am curious to know is the adhesion of the die onto the tab(paddle). The tab is coated with a Ag layer. How does the gold layer (from the preform) actually form adhesion to the silver plated frame. Is this by diffusion? Does Au diffuse easily into the silver layer? (The melting point of Au as I understand is 1064C; so how does it form an adhesion with silver? Our die bond temperature is usually less than 450C) Would appreciate your help. Sorry I am not an assembly engineer. I am just curious and would appreciate any help. Thanks and regards. |
|