Hi Yu,

I think your RIE process is isotropic. What is the feature size of your
silicon bridge? Undercutting usually severes during overetch time. You
can check whether your Cr etchant eats your silicon by doing a profile
scan after Cr wet etch to determine the thickness. If it is significantly
thicker than your expected Cr thickness, your silicon must be partically
etched. For anisotropic RIE process, you should use CF4/H2 or CHF3, and
try pressure below 10 mTorr. But watch for polymer residues.

Yours sincerely,

Isaac Chan
University of Waterloo
--------------------------------------------------------------------
My experience is on a-Si:H TFT design and semiconductor processing:
http://resumes.hotjobs.com/iwchan2004/processengineer6yr
--------------------------------------------------------------------

On Mon, 10 Jan 2005 cyyfq@bu.edu wrote:

> a question about RIE undercut.
> I am using RIE to etch silicon(O2/CF4=5/50 sccm, pressure 50 m torr, power 300
> W), wafer is SOI wafer, devise layer 230nm,100 direction
> the mask is 100nm of Cr.
> after etching , I use Cr etchant( bought from Transene company, LTD, type
1020)
> to remove Cr mask.
> It seems that there is a huge undercut around 100nm,
> ( I think so because the silicon bridge does not look sharp and flat, and
> resistance is much bigger than expected)
> I  etched only 230 nm deep, this is quite unresonable,
> it looks like the side of the bridge was peeled off  a little bit.
> there might be three possiblitis for this problem.
> 1. SOI wafer is not good.
> 2. RIE under cut( pressure has be set  very low already).
> 3. Cr etchant eat some silicon.
> anyone have some idea of this problem and how to solve it.
>
> thanks a lot
> Yu Chen
> cyyfq@bu.edu