hello carsten,

in case of particles during the process, there can be a lot of route causes.
improtant is to find out where the particles are from.

before doing any checks, please make sure that the machine itself is in good
conditions! check all values for vacuum, leak rate, pressure rate, bias,
power, etc. you should also check the lifetime of all chamberparts. if the
lifetime of any part is out of spec, exchange this part as it is recommended
from the equipment supplier.

=> if everything that you checked is ok, i would recommend to do the
following procedure:

1. if easily possible: do a particle analysis to make sure of which
materials the particles consist. with those informations you are sometimes
able to localize the weakpoint directly (sealing, filtermaterials, etc.)

2. in any case: do detailed handling checks without using process gases. =>
a) check only loadlock, b) loadlock to orienter (if orienter exists!), c)
loadlock to transfer chamber and back to loadlock, d) loadlock to process
chamber and back to loadlock (without uising gases! check only every robot
moving and robot handling by moving the wafer into process position and pump
down the chamber!)

3. if step 2 was not successful, do handling checks from loadlock to process
chamber and back, checking every gasline in single mode => important: for
every gasline, use a new check-wafer!

!!!important for all particlechecks: pre measurement => transfercheck =>
post measurement!!! even if you use brand new wafers for the check, do the
pre measurement, that is my experience from 4 years of responsibility for
productionprocesses and -equipments in a semiconductor massproduction. often
you get still somehow "dirt wafers", especially, if you are using recycled
wafers for those checks.


there is also one other possibilty for your particle problem:
could it be, that the freuence of the very intensive chamber cleaning is to
high?! you can get some kind of overetch, so that chamber parts are attacked
by the cleaning by and by, so that there are particles generated. i think
you have to think about that... => i would do a intensive manual wetclean of
the chamber and then start with a chamber that is in absolut best
conditions. perhaps think about an extension up to every 2, 3, 4 or 5 wafer
clean, then do an intnesive cleaning and process next wafers...to prevent a
so called first wafer effect, simulate the first wafer or use a dummy wafer
as the first wafer.

anyhow, i wish you a lot of luck and hope that you will have success finding
the particlesource... :o)))

best regards,
michael
(m.proempers@fz-juelich.de)




> -----Ursprüngliche Nachricht-----
> Von: mems-talk-bounces@memsnet.org
> [mailto:mems-talk-bounces@memsnet.org] Im Auftrag von Weber,
> Carsten - LT
> Gesendet: Donnerstag, 14. September 2006 15:03
> An: mems-talk@memsnet.org
> Betreff: [mems-talk] Small black particles falling down
> during SiO2 dry etch in AMS200
>
> Dear all,
>
> I just want to know if there is anybody who has a similar problem.
> When I etch 2300 nm of thermal SiO2 in our AMS200 ICP
> Dielectric Etcher
> (with use of C4F8 and H2) I got problems with small particles 1 micron
> big for example which fall down during process an passivates
> my etching
> structures.
> I think this particles are coming from the aluminium clamp
> ring shield.
> After each wafer we make a O2 Cleaning because we are etching with
> C4F8. Cleaning is as long as one wafer will be etched (approximately
> 7,5 minutes).
>
> My process is like this:
>
> C4F8: 13 sccm
> H2:   40 sccm
> pressure: 2,8 E-3 mbar
> Source:       2800 Watt
> Substrate Holder: 250 Watt
> Temperature: 0°C
>
> Is there maybe a possibility for another process with good
> etch rate and
> good selectivity to PR?
> On our ICP we have this gases available:
>
> CF4, CHF3, H2, C4F8, O2
>
> I use PR-Mask (1,7 microns thick) to etch 2200 nm thermal oxide.