I have seen the same effect with Gold on top of Nickel, and using an Iodine
based gold etch.  Doesn't normally affect Nickel, but with this arrangment,
the nickel etched extremely rapidly.

I had a vague theory that there was a galvanic potential caused by the
metalic interface causing the nickel to act as an anode and electroetch
itself out into solution.  But I never figured it out, and don't know nearly
enough chemistry to know if this makes any sense.  If you get a compelling
answer, I'd love to hear about it.

David Nemeth
Senior Engineer
Sophia Wireless, Inc.
14225-C Sullyfield Circle
Chantilly, VA 20151


-----Original Message-----
From: MEMS R FUN
Sent: Monday, August 22, 2005 2:09 PM
To: [email protected]
Cc: [email protected]
Subject: [mems-talk] Selective Cu & Cr Etchant


All,

I was wondering if anyone knows of an etchant that will eat through layers
of copper and chromium without etching nickel.  I have a simple process by
which I am trying to electroplate nickel over a Cr/Cu seed layer
(200A/1000A) into thick PR (SPR 220-7) patterned spots.  Post-plating, I
would like to remove the PR layer and etch away the exposed seed layer to
give me just the nickel and silicon substrate.

My first attempts at this were done using CR-7.  According to K. Williams'
etch table (JMEMS Dec 2003), CR-7 should etch copper at a rate of 280
nm/min, chromium at a rate of 170 nm/min, and nickel at a rate of 1.7
nm/min, with all of the metals above being evaporated.  HOWEVER, when trying
this experimentally in a diluted solution of CR-7 (50% by vol DI), the
copper was etched away (as expected), the nickel was RAPIDLY etched very
much unexpected), and ironically enough, the chromium was not etched (or was
etching extremely slowly...again, very much unexpected).

Is it possible that the Cr and Cu formed intermetallics at the interface
that would prevent etching or did something else happen to impede the
etching using CR-7?

Is it possible that the nickel etch rate could increase by several orders of
magnitude if the nickel is electroplated as oppsed to being evaporated?  My
thoughts are no, but I'm open to hear other opinions since my experiments
have proven otherwise.

Does anyone know of another recipe/etchant that will work?