A MEMS Clearinghouse® and information portal
for the MEMS and Nanotechnology community
RegisterSign-In
MEMSnet Home About Us What is MEMS? Beginner's Guide Discussion Groups Advertise Here
News
MEMSnet Home: MEMS-Talk: Re: deep Si etching
Re: deep Si etching
2001-02-23
Nickolay Lavrik
2001-03-07
Nickolay Lavrik
Re: deep Si etching
Nickolay Lavrik
2001-03-07
--------------CC2430D150706E69DBEEF057
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> that's a really great way for KOH masking.
>   could you mind share PECVD nitride recipe as KOH mask with us?
>   thank you very much.
>
It looks like there is a considerable interest in making KOH resistant
PECVD nitride.

Before going into the details I want to remind that PECVD is a rather
"dirty" process and only in very few cases it is worth being considered
as an alternative to LPCVD . Not to mention that PECVD nitride can NOT
be used for making free standing membranes.

In attempt to prepare KOH resistant PECVD, I was varying the regimes and
was trying some post-treatments.  Eventualy I ended up with a couple of
samples that did not undergo any etching. These were the samples
prepared at the highest possible substrate temperature (it was 400-450 C
for the chamber I used), while the power, the pressure and the ratio of
SiH4/NH3 gases
were quite typical for PECVD nitride  (the exact settings will depend on
a
specific apparatus). In addition,  these samples were annealed, after
PECVD deposition of nitride, in oxygen atmosphere at low pressure at 450
C for 25-30 minutes.  This annealing was done inside the PECVD apparatus
using "oxygen cleaning" settings but without initiating the plasma (RF
power=0). In fact, one of the samples did undergo "oxyden cleaning" with
the plasma ON for 20 min and was KOH resistant as well. Some other
samples with PECVD nitride were annealed simply on a ceramic-top hot
plate in air at 500 C  (a rough estimate) for 5-15 min and also appeared
to be extremely stable during  KOH etch.

Should any of you try similar procedures, I would be interested to know
how reproducible my observations are.


Nickolay

--------------CC2430D150706E69DBEEF057
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit




that's a really great way for KOH masking.
  could you mind share PECVD nitride recipe as KOH mask with us?
  thank you very much.

It looks like there is a considerable interest in making KOH resistant PECVD nitride.

Before going into the details I want to remind that PECVD is a rather "dirty" process and only in very few cases it is worth being considered as an alternative to LPCVD . Not to mention that PECVD nitride can NOT be used for making free standing membranes.

In attempt to prepare KOH resistant PECVD, I was varying the regimes and was trying some post-treatments.  Eventualy I ended up with a couple of samples that did not undergo any etching. These were the samples prepared at the highest possible substrate temperature (it was 400-450 C for the chamber I used), while the power, the pressure and the ratio of SiH4/NH3 gases
were quite typical for PECVD nitride  (the exact settings will depend on a
specific apparatus). In addition,  these samples were annealed, after PECVD deposition of nitride, in oxygen atmosphere at low pressure at 450 C for 25-30 minutes.  This annealing was done inside the PECVD apparatus using "oxygen cleaning" settings but without initiating the plasma (RF power=0). In fact, one of the samples did undergo "oxyden cleaning" with the plasma ON for 20 min and was KOH resistant as well. Some other samples with PECVD nitride were annealed simply on a ceramic-top hot plate in air at 500 C  (a rough estimate) for 5-15 min and also appeared to be extremely stable during  KOH etch.

Should any of you try similar procedures, I would be interested to know how reproducible my observations are.
 

Nickolay --------------CC2430D150706E69DBEEF057--

reply
Events
Glossary
Materials
Links
MEMS-talk
Terms of Use | Contact Us | Search
MEMS Exchange
MEMS Industry Group
Coventor
Harrick Plasma
Tanner EDA
Addison Engineering
Process Variations in Microsystems Manufacturing
The Branford Group
University Wafer