Crater defects are usually a result of TMAH or KOH etch on wafers with
inverted etched  pyramids  defects to start with.
If your dopant removal can cause these type defects - this may be the
reason.
Shay Kaplan

-----Original Message-----
From: [email protected] [mailto:[email protected]]
On Behalf Of Brian Stahl
Sent: Thursday, October 16, 2008 12:03 AM
To: General MEMS discussion
Subject: [mems-talk] Crater Defects in Silicon Wet Etching Process

I'm finding crater defects in heavily boron-doped silicon when I etch
through the wafer with 25% TMAH at 85C.  These defects seem to come from the
spin-on dopant/thermal diffusion process I use to dope the wafers, as I
don't see these defects in undoped wafers and the defects are larger and
more numerous in wafers that had longer diffusion times.  These craters are
perfectly smooth, concave surfaces that are depressed into the (otherwise
flat) surface of the etch cavity.

Has anyone come across these defects before?  They are not the same as the
hillocks commonly associated with TMAH etching at lower concentrations.  My
hypothesis is that they're volumetric defects created during the doping
process, either from impurities from the spin-on dopant (hydrogen?) or
impurities/defects that congregate from the bulk silicon.  I can observe
them throughout the entire thickness of the wafer, not just near the heavily
boron-doped region.

Any advise you can offer would be greatly appreciated.

Thanks,

Brian C. Stahl