Sputter is not the ideal choice for lift-off metallization.

Sputter tends to be relatively conformal.  And, the metal coating on the
sidewalls of the lift-off resist will work as a bridge between the metal
on the substrate and that on the top of the resist.

In order to have successful lift-off patterning, you need as much
separation between the metal on top of the resist and that on the
substrate.

1 - Select a metal process that has poor step coverage.  Evaporation is
best, but sputter may work if you adjust it a bit (maybe try lower
pressure or a longer throw distance?).

2 - The resist sidewalls should be negatively sloped (overhanging) as
much as possible.

3 - The ratio of resist height and metal thickness should be as large as
possible.  I think I remember reading that 10:1 is ideal.

Good luck,
Dan Ruiz
MEMS process engineer
Honeywell
15001 NE 36th Street
Redmond, WA  98073-9701
Phone - 425-376-2167
[email protected]

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf
Of ??
Sent: Monday, May 23, 2011 5:32 AM
To: [email protected]
Subject: [mems-talk] metal lift off problem

Hi,

     I am using the e-beam to pattern my sample,
the resulotion is not very high, just around 500nm, and my
sample is 10nm aluminum on sapphire.

    It is followed by sputtering 100nm metal aluminum.
    After that, I use the aceton to do the lift-off.

    My problem is that the metal did not peel off as expected. I
appreciate any advice.

    Thank you!