Use APTES as diffusion barrier layer.  The secret is to promote total
dehydration then APTES as a vapor can react with all exposed areas.
Resulting APTES barrier even though only 20 Angstroms thick stops all
copper diffusion into Polyimide layer.  Contact me directly we
manufacture a Polyimide bake unit that bakes the polyimide correctly and
we have now incorporated a APTES vapor delivery system onto this bake
unit.


Bill Moffat, CEO
Yield Engineering Systems, Inc.
2185 Oakland Rd., San Jose, CA  95131
(408) 954-8353

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Michael D Martin
Sent: Friday, February 10, 2006 1:27 PM
To: [email protected]
Subject: Re: [mems-talk] diffusion barrier for copper-polyimide

Erikin,
   You might try using the polyimide adhesion promoter, VM-651 which I
think will stick to metal oxides.

-Michael
  U. of Louisville

>>> [email protected] 02/10/06 1:39 PM >>>
Hi,

I am trying to make micro strain gauges using the following method in
summary:

- Spin 250nm-thick polyimide diluted with NMP; completely cure at 325C
- Sputter/pattern 80nm-thick Cu/Ni on cured polyimide
- Spin 250nm-thick polyimide diluted with NMP; completely cure at 325C

The problem I am having is that after the cure, the polyimide above
Cu/Ni layer is disappearing.  I read about copper oxide diffusion into
polyimide due to the reaction between copper and polyamic acid; but I am
not sure if this explains the material loss over the metal lines.

I already tried amino silane adhesion promoters provided by GE Silicone,
but didn't have any success.

I would be grateful if somebody could suggest a solution.