This cratering is not exclusive to KOH etch chemistries - I observed the
same defects when etching ~500um deep square wells with TMAH.  I have also
observed them on commercial AFM cantilever chips.

I never conclusively determined the cause, but I strongly suspect the
cratering is due to a crystallographic defect in the silicon - either
dislocation cores or voids that form from aggregation of point vacancies.
 There is an activation energy for vacancy formation and their equilibrium
concentration is therefore temperature-dependent.  During cooling from a
high-temperature annealing or oxidation process the vacancies generally
don't escape to a free surface, so they aggregate and form voids; these
voids can also form during ingot cooling in the Czochralski process.

I hope this helps.

Best regards,

Brian Stahl, Ph.D.
Apeel Sciences


On Thu, Jul 3, 2014 at 4:41 AM, Sales  wrote:

> Hi Aaron.  The only potential cause that comes to mind in this case would
> be some residue from the stripping process.  Perhaps some areas are masked
> a little and so protected to an extent while others are more exposed and
> form the craters.  This is not normal behavior for KOH solutions--they
> should etch nicely along the crystal plane.
>
> Incidentally, we do offer KOH solutions if you want pre-made product.  I
> do not think your results would be any different, but if you determine the
> cause we can help you with the chemistry.
>
> Regards,
>
> Christopher Christuk
> Transene
>
> -----Original Message-----
> From: [email protected] [mailto:
> [email protected]] On Behalf Of Aaron
> Glatzer
> Sent: Wednesday, July 02, 2014 1:42 PM
> To: [email protected]
> Subject: [mems-talk] KOH etch surface "craters"
>
> We are etching some 300 micron deep trenches using KOH, and we are
> occasionally getting "craters" in the trench bottom surface.  These are 2-4
> micron deep 50-75 micron wide spots on the bottom of the trench.
>
> Is this typical for KOH etch?  Has anyone else seen this kind of thing?
>  Any idea what causes it?
>
>
> Technical details:
>
> <100> silicon.
>
> Masked with 2kA low stress LPCVD nitride.
>
> Nitride etched with SF6 plasma RIE; we guesstimate we etch 2000A into the
> Si
> (based on known-etchrate, and observed overetch time on endpoint signal).
>
> Resist stripped in O2-ash/piranha/O2-ash sequence.
>
> KOH 45%, etched at 85C, for about 5.5 hours.
>
>
> Typically we get a mildly textured surface (per optical inspection), but as
> indicated above occasionally we get what look in microscope like blisters,
> but optical profilometer indicates is actually a "crater".
>
> Any insight into this is appreciated.
>
> thank you,
> Aaron G.
>
>
>
>
>
>
> --
>
>
> Aaron Glatzer                                           phone (419)
> 241-6963 x13
> Lead Process Engineer                           fax (419) 241-6966
> Midwest Microdevices, LLC
> [email protected]
> 329 14th Street
> Toledo, OH 43604
>
> ---
> This email is free from viruses and malware because avast! Antivirus
> protection is active.
> http://www.avast.com
>
> _______________________________________________
> Hosted by the MEMS and Nanotechnology Exchange, the country's leading
> provider of MEMS and Nanotechnology design and fabrication services.
> Visit us at http://www.mems-exchange.org
>
> Want to advertise to this community?  See http://www.memsnet.org
>
> To unsubscribe:
> http://mail.mems-exchange.org/mailman/listinfo/mems-talk
> _______________________________________________
> Hosted by the MEMS and Nanotechnology Exchange, the country's leading
> provider of MEMS and Nanotechnology design and fabrication services.
> Visit us at http://www.mems-exchange.org
>
> Want to advertise to this community?  See http://www.memsnet.org
>
> To unsubscribe:
> http://mail.mems-exchange.org/mailman/listinfo/mems-talk
>
_______________________________________________
Hosted by the MEMS and Nanotechnology Exchange, the country's leading
provider of MEMS and Nanotechnology design and fabrication services.
Visit us at http://www.mems-exchange.org

Want to advertise to this community?  See http://www.memsnet.org

To unsubscribe:
http://mail.mems-exchange.org/mailman/listinfo/mems-talk