Hi there,

We have done the following experiment to find out which is the
fastest etching plane:
We took an anisotropically etched trench with etch mask removed
(thus exposing a protruding <111>-<100>-corner) and etched it
in the same solution again. The protruding corner will be cut
by whatever is the fastest etching plane (provided it etches
faster than <100>).
In 5% TMAH at 85 deg C, this was very clearly <311>, not <411>,
but the result is very dependent on etchant, concentration and
temperature - maybe that explains the variations in literature?
This method does not allow accurate measurement of the etch
speed in that plane.

Greetings,
        Frank Berauer
        Senior R&D Engineer
        Hewlett-Packard Singapore


-----Original Message-----
From: Roger Shile [mailto:rshile@tmmicro.com]
Sent: Tuesday, April 23, 2002 5:26 AM
To: mems-talk@memsnet.org
Subject: Re: [mems-talk] Convex Corner Compensation


Here are a couple of references on crystallographic etching and compensation
of convex corners on silicon:

Mayer et. al, J. Electrochem. Soc. Vol 137, No. 12 (Dec '90) p. 3947

Offrens et. al. Sensors and Materials, 3,3 ('92) p. 127

Zubel, Sensors and Actuators A 94 (2001) p. 76

The convex corners of a mask tend to be undercut by fast etching planes.  I
am
puzzled by the disagreement in the literature as to which planes etch
fastest.
The first two references above indicate that in KOH etching of Si, the the
highest etch rate always occurs normal to the <411> plane.  Data given by
the
third reference indicates that the highest etch rate occurs normal to the
<311> plane.

The ACES etch simulator gives the <311> plane the highest KOH etch rate and
doesn't even allow specifying an etch rate for the <411> plane.

The ratio of etch rates on different planes is a function of the etchant
composition (e.g. KOH vs. EDP), the concentration (e.g. 20% KOH vs. 40% KOH)
and the temperature.

I would be grateful is someone could explain the reason for the apparent
contradiction in the literature as to which is the fasting etching plane
when
Si is etched with KOH.

Knowledge of which planes are dominating the undercut of convex corners is
necessary to properly design corner compensation and other masks for
processes
which utilize this undercut.

Roger Shile


>>> michelefinardi@yahoo.it 04/22/02 01:48AM >>>
Hallo,
I'm about to start working on convex corner
compensation. Could you please indicate me some
references or share your personal knowledge?
Thank you very much.

Michele Finardi

______________________________________________________________________
Vinci i Mondiali - La Grande Sfida dei Mondiali FIFA
http://it.yahoo.com/mail_it/foot/?http://lasfida.yahoo.net/
_______________________________________________
mems-talk@memsnet.org mailing list: to unsubscribe or change your list
options, visit http://mail.mems-exchange.org/mailman/listinfo/mems-talk
Hosted by the MEMS Exchange, providers of MEMS processing services.
Visit us at http://www.mems-exchange.org/
_______________________________________________
mems-talk@memsnet.org mailing list: to unsubscribe or change your list
options, visit http://mail.mems-exchange.org/mailman/listinfo/mems-talk
Hosted by the MEMS Exchange, providers of MEMS processing services.
Visit us at http://www.mems-exchange.org/