I was looking for anyone who may be able to offer some advice on the use of
and effectiveness of RCA standard cleans versus piranha cleans. What I am
evaluating now is the feasibility of changing over to RCA cleaning from our
current sulfuric peroxide clean process. The reason for the interest in
making the change involves the possibility of an individual using KOH etchant
followed by the use of some of our furnaces that are normally kept free of
MEMS processing cycles, to remain CMOS-ready.
We have used piranha cleans for a number of years (before I came to the lab)
and contamination problems haven't cropped up before, so there was no
apparent need to change anything. However, without solid data on surface
contaminants, the adequacy of H2SO4/H2O2 was not truly determined. I applied
the "If it ain't broke, don't fix it" mantra, using available information.
Now, however, we are worried that unless a very thorough metal clean is
performed on wafers following KOH etch (as stressed by other CMOS/MEMS
integration labs), we may run into contamination problems. I haven't looked
over comparison studies on the single clean (sulfuric peroxide) versus the
two step RCA, but had a feeling that the RCA was more efficient, as the
P-clean was an optional step additional to RCA, and had not heard of this
used as a standalone, for the most part.
What things come down to now is in trying get any feedback people may wish
to provide on setting up/maintaining RCA cleans. If possible, I need to look
at a conversion on the existing station over to RCA from a two bath sulfuric
peroxide (resist strip and clean) with hot/cold DI rinses to RCA-1/RCA-2. I
understand that the typical process is 5:1:1 H2O/NH4OH/H2O2 @ 80C, 6:1:1
H2O/HCl/H2O2 @ 80C, with HF dip between for silicon/polysilicon, and using
cold DI rinses to resistivity. (Research shows much weaker solutions to be as
good or better with proper maintenance, but that is another issue). I am also
aware of a possible particulate problem if exhausting both RCA-1 and RCA-2 in
the same hood, due to NH4Cl formation. Again, the station we currently use was
not designed with RCA cleans in mind, and I am trying to get any feedback from
the manufacturer if possible.
Some issues that came to mind are with regards to:
1) frequency of bath changes, and monitoring (loss of) effectiveness of the
clean cycle.
2) injection of new H2O2 into baths to make up for depletion (We do this now
with a fitted pump in the H2SO4).
3) purchase of and any mixing of solutions (pre-mixed, mixing on site?)
4) any special concerns of RCA over sulfuric peroxide use.
Thank you to anyone who wishes to provide any feedback to this information.
John
-- ---------------------------------------------------------------- John M. Koons *NJIT Microelectronics Research Center* Research Engineer e-mail: jmk4336@megahertz.njit.edu phone: (201) 596-8450 fax: (201) 642-4848 ----------------------------------------------------------------
This archive was generated by hypermail 2b29 : Tue Mar 09 2004 - 07:48:59 EST