KOH and Gold Question

Kris Vossough vossough at snf.stanford.edu
Tue May 22 01:03:10 PDT 2001

Hi Moe,

I struggled with similar situation in the past. Generally, good adhesion
of the Gold layer depends on the stability of the adhesion layer,
Chromium. In order to maintain good adhesion between Cr and Au layers, the
predeposition (base) pressure for Cr/Au must be low, i.e. 1 to 4E-7 torr.
Otherwise, the deposited Cr oxidizes rapidly and will not adhere to the Au
film (contaminated surfaces have similar effect). I have been able to
extend the life time of my Cr/Au films in BOE solution for up to 1 hour.

I am not sure about adhesion to the nitride layer, but I suspect rougher
surface on the nitride film (or other geometrical factors) may be the
reason. Ti can also be used for adhesion layer, but it oxidizes even
faster than Cr.

Kris Vossough
TelOptics Inc.

On Mon, 21 May 2001, Mohammed H. Badi wrote:

> Hi all,
>    I'm doing an 80C KOH etch of silicon and have been having some trouble.  I
> have chrome/gold (100A and 3000A respectively) pads on my wafer that are 90um
> x 90 um that REFUSE to stick to the silicon underneath it.  The pads (and
> thin gold lines) do, however, stick to silicon nitride on other parts of the
> wafer.
>    My question for you: is there anything I can do to stop the gold pads from
> floating off?  Maybe a different adhesion layer?  A different etch 
> temperature? 
>    These gold squares on silicon act as ground pads for my devices.
>    Thank you.
> moe.			
> --------
> Mohammed H. Badi		"That's just so typically me."
> 650-906-0663	

More information about the wbgen-hpl mailing list