| Cleaning is not an easy thing for those working at the nanoscale
and in the semiconductor field. In fact, many simple approaches to
cleaning
will result in a "dirty" surface for our needs. Alas, others
have reached and overcome this hurdle in many ways: RCA Clean A well-known and established technique, RCA clean consists of two
to three steps. This was developed by Kern in 1965. These chemicals are dangerous! Be careful, wear full safety gear,
do not store it in a sealed container, and check with your safety
department on how to properly dispose of it before mixing it up.
[1] W.
Kern, J. Electrochem. Soc., 137, 1887 (1990).
[2] Handbook of Semiconductor Wafer Cleaning Technology:
Science, Technology, and Applications, W. Kern, Editor,
p. 433, Noyes Publishers, Park Ridge, NJ (1993).
The basic process, from [2],
is:
- [Remove organics
and many metal contaminants] RCA Standard Clean 1 (SC1)
- 5:1:1
parts by volume of
nanopure water (NPW), NH4OH (29% w/w% as
NH3),
and H2O2 (30%, not stabilized).
- Maintain solution
at 70-80° C
(stay within this range) for 5 min.
- Overflow quench with
water (pour in water so that
the liquid overflows, so as to displace the now-contaminated
surface layer of liquid)
- [Optional step* - oxide etch,
typically used with silicon wafers] HF Etch
- 1% HF:H2O (e.g.,
1:50 if using 49% HF) x 15 sec
- Rinse with NPW
- [Remove alkali ions,
NH4OH-insoluble hydroxides, and residual trace metals] RCA
Standard Clean 2 (SC2)
- 6:1:1 parts by volume of NPW,
H2O2 (30%, not stabilized), and HCl (37 w/w%)
- Maintain a solution temperature of 70° C for 5-10 minutes.
* - this step can be performed
first; check you SOP and read Kern's book if you are unsure if you
should use it and when. Some online SOPs:
A modification is described here: Celler
et al, Electrochemical
and Solid-State Letters, 3 (1) 47-49 (2000). Piranha Clean (removes organic contaminants) This refers to an acid-peroxide mixture that gains its name from its
aggressive etching of organic contaminants. Needless to say, be careful
around this.
I use the following protocol
from Kern's book on Wafer Cleaning:
- Maintain 2:1 ratio of
H2SO4:H2O2 at 130 C x 15 minutes.
- Note: this is a highly exothermic
reaction. The temp will heat up to about 100-110 C on its own within
miutes, then it will start to drop after 4 -5 minutes, so maintaining
130 C can be tricky.
- Again, this is a very
dangerous procedure. You should strive to have an experienced person
train you thoroughly and use ample safety gear (heavy gloves, safety
glasses and/or face shield) and work in a hood.
- The mixture may fizz
out of your container, so assume your gloves/hands are covered
with piranha solution regardless, and act accordingly.
Some personal technique notes (when cleaning pieces of a wafer):
- I really do strive to keep the heat at 130 C or above for the full
15 minutes; I have seen some levels of successful cleaning without
worrying about
this
(mix
the reagents without additional heating), but when cleaning diced
sapphire wafers it was essential to maintain the target heat (although
this was tricky and it ranged as high as 150 C for small periods
of time).
- Teflon dipping basket - I aim for zero contact between tweezer
and sample in the piranha solution. I place my wafer piece in a pre-cleaned
Teflon basket (~50 mL Teflon beaker with holes drilled into it and
a crude handle made from Teflon-coated wire), which I then use to
introduce and remove the samples. The tweezers do not
touch
the
substrates
again
until
after
the piranha
has been rinsed off by copious amounts of water.
- Use lots of rinse water - don't be shy, you want it clean. When
I pull the small basket (above) out of the reaction, I pour about
500 mL of 18 megaohm nanopure water (NPW) through/over the
basket/samples. I then pull them out one at a time, rinsing them
with a wash bottle
of NPW for about 10 seconds a side.
- Dry well - to help insure a clean surface, I dry the chips immediately
after rinsing. I use a dry nitrogen from a cylinder passed through
a pipet tip that is trimmed to give my
desired
air flow.
UV/Ozone Exposure (removes organic contaminants) Simply put, you expose the sample to a UV/Ozone environment. There
are cheap ways to accomplish this crudely, e.g. an appropriate UV penlight
in a box, or you can buy a system that is plug and chug with a timer,
tray,
and
such. The time you use will depend on your application; I usually went
with usually went with 30 minutes after HF etching silicon wafers.
Just google
for "UV ozone cleaner". HF Etch (etches silicon and its oxide)
- This stuff is
dangeorus; it goes straight to the bone! Be careful, wear
full safety gear, do not store it in a sealed container,
and
check with your safety department on how to properly dispose
of it before mixing it up.
- HF will race to your bones and supplant the calcium with
a toxic, insoluble substance, and cause liver failure. It
takes a surprisingly small amount. It is wise to have clacium
gluconate on hand to treat spills.
- One of the steps
used in the RCA, but also a viable technique on its own.
- 1% HF:H2O, e.g.
with 49% HF dilute 50:1 H2O:HF, etches SiO2 at a few angstroms
per second. It is common to use anywhere from 1% to 5% mixtures.
Aqua Regia (dissolves gold and platinum)
Aqua regia, which translates
from Latin into "royal water", earned its name due to its
ability to dissolve
"royal" metals such as gold and platinum. The components,
HNO3 (nitric acid) and HCl (hydrochloric acid), are typically mixed
in a 1:3 ratio
of HNO3:HCl. Interestingly, the components do not appreciably dissolve
the metals on their own.
- This stuff is dangerous! Be careful, wear full safety
gear, do not store it in a sealed container, and check with your safety
department on how to properly dispose
of it before
mixing
it up.
- Reverse Aqua Regia, a
mixture of 3:1 HNO3:HCl can have special applications, which are
discussed briefly on this
site.
Last updated
2009-07-30
|