durusmail: mems-talk: Cl- and Br-contained plasma - toxic?
Cl- and Br-contained plasma - toxic?
Cl- and Br-contained plasma - toxic?
Martin.WALKER@oxinst.co.uk
2003-02-18
Hi Isaac, everyone

There are some very toxic gases used for plasma etching, with special
precautions needed for certain gases.  Here is a general summary.

Fluorine containing gases:  Some are dangerous out of the cylinder (such as
NF3) and need special precautions with new pipework, etc.  Many are inert
before they go into the plasma (CF4, CHF3, SF6) but will release F2 or HF in
the plasma.  This will go into the exhaust.  If you are using wet pumps,
this will dissolve in the oil.  The oil (and any other residues in the
exhaust system) should be treated in the same way as hydrofluoric acid, that
is, with great care.  Exhaust systems should either remove this using a
suitable absorber, or should dilute and vent to atmosphere well away from
sensitive areas (exact details depend on the country/state/county you live
in).  First aid precautions should include calcium gluconate gel for dealing
with HF burns.  First aiders should be trained in how to deal with such
injuries.

Chlorine and Bromine containing gases:  These would not normally be used in
open-load plasma chambers, due to the toxicity of the gases and of the
residues remaining on the chamber walls.  If they are being used in a
non-load-locked system, a glove box should be used to protect the user from
these residues, which are produced when moisture reaches the process
chamber.  It is normal to have a fixed point continuous monitoring system
when these gases are in use.  Most sensors which are tuned for Cl2 or HCl
have cross sensitivity for Br compounds too.  Fortunately, all of these can
be smelled at very low concentrations, so are not as great a risk as gases
such as PH3.  Where-ever possible, metal to metal seals should be used
(including in mass flow controllers).  Where o rings are used, these would
normally be Kalrez rather than Viton.  Again, exhaust gas treatment or
dilution and controlled venting should be used, with the exact precautions
depending on the quantities and the local legislation.

For all plasma etch systems, the most hazardous operation is chamber
cleaning.  Even if the feed gases are inert and the chamber has been pumping
for a number of hours, toxic compounds can come off the walls during
cleaning.  Often, either dust masks or even full-face respirators are
recommended protective equipment for this operation.  A full risk assessment
should be done before carrying out such operations.

Before using (or installing) any gases, consult the relevant Material Safety
Data Sheet (MSDS).  These should always be supplied by the gas supplier.  If
they haven't supplied them, demand them!  Most are pretty good, but
sometimes they err on the side of caution (usually to minimise the chance of
litigation in the US).

The semiconductor market (and the MEMS market as a spin-off) generally has a
very good safety record.  Much of this is because people are very willing to
share safety related information.  If you have questions about safety, never
be afraid to ask!

Regards,

Martin

Martin Walker B.Sc.(Tech.) M.Sc.
Tactical Marketing Engineer
Oxford Instruments Plasma Technology
North End, Yatton, Bristol, BS49 4AP UK
T.+44 (0)1934 837031  F.+44 (0)1934 837001
E. 
W. 


>Dear all,
>
>       I have another question regarding plasma etching (please see
>email "Subject: post-RIE photoresist residue" for the other question).
>Besides, CF4, CHF3, C2F6, and other fluorine-contained plasmas, Cl- and
>Br-contained plasmas, such as CF3Cl and HBr, are commonly used for
>dry etching. However, are they very toxic and need special chamber, pump,
>and safety conditions to use those gases for dry etching? The thin film
>materials for etching are amorphous silicon, amorphous silicon nitride,
>molybdenum, and (maybe) chromium. Any comments are appreciated.
>
>
>Isaac



 ###  OXFORD INSTRUMENTS   http://www.oxford-instruments.com/  ###

Unless stated above to be non-confidential, this E-mail and any
attachments are private and confidential and are for the addressee
only and may not be used, copied or disclosed save to the addressee.
If you have received this E-mail in error please notify us upon receipt
and delete it from your records. Internet communications are not secure
and Oxford Instruments is not responsible for their abuse by third
parties nor for any alteration or corruption in transmission.


reply