Chris,

How about using KI/I for the gold wet etch? This is a pretty selective etch. I
don't believe it attacks aluminum, but even if it does, it is a lot slower than
aqua regia.

Jordan

On Apr 4, 2011, at 11:00 AM,  "Christian Engel" wrote:

Date: Mon, 4 Apr 2011 00:09:48 +0200
From: "Christian Engel" >
Subject: [mems-talk] Contact of 2 Al-Layers
To: 

Hello everybody,

I have 2 substrates, each with structured metal-layers. I want to bond the
substrates together. There are Pads on each surface for an electric contact
between the two metal-layers. The pads are a little bit recessed in
cavities. First we tried electroless plating for forming the
metal-structures but soon we faced heavy problems that aren't solved yet. So
we tried thick PVD Aluminium and thin PVD Gold on top as oxidation barrier.
Wet etching of this metal-combination failed because etching Gold in aqua
regia strongly attacked the Aluminium. Because time is running we decided to
use aluminium only because we can handle this material without problems.
Except its problematic native oxide. I don't really have an idea how to
avoid electrical contact problems because of this oxide. Even if I do a
short etch-dip for removing the oxide I am afraid that evacuating the bonder
takes too much time and the oxide will be there again before bonding starts..
Perhaps anybody has a great idea about how to handle this sorrowful
situation. Perhaps I can use some kind of conductive glue and get the
substrates together without bonding? Unfortunately the process needs a real
bond because there will be cavities between the two substrates that should
be "filled" with vacuum...

Thanks a lot and best regards, Chris