>Dear Isogeochemers,


We suck out the ash from the furnace on the EA with a stainless steel tube
connected to two different flask that remove the large and then small parts
of the ash.  Then the air goes through a GAST 23 series pump with 10 micron
filters onthe inlet and outlet, and is then vented into a fume hood.  We do
not attempt to suck out al the ash down to the level of  the quartz wool,
but leave about 5-10mm of ash on the surface of the quartz wool.  We have
the stainless steel tube marked so that we know how far into the combustion
tube we are.  This ash layer rapidly becomes hard and seems to protect the
quartz wool, and chromium below.

We usually suck out the ash at the start of the working day, and then run
up to 200 organic samples through the instrument before the next morning
when we would suck out the ash again.  For soils we would only run 60 soil
samples before sucking out the ash.  We have found that leaving soils in
for any length of time results in them forming into a large solid block of
ash that cannot be removed. Therefore the rule here is that soil ash is
removed immediately after the run or first thing the next morning.  I will
bring photographs of the suction system and a stainless steel tube to the
Canadian conference in August.

Out of interest we use 100 mL min-1 Helium flow rate with an analysis time
of 6 minutes per sample on the Europa Sci. 2020 and 8.3 minutes on the
Carlo Erba 1500/Delta XL system.  In practice throughput of the unknown
samples is similar on each system as we have to run more standards with the
Europa system since it doesn't have reference gas injection.

We usually use 9x5 tins for all samples, and can still get up to 200 plant
material samples as heavy as 15mg each burned without sucking out the
ash.  Presumably with organic samples most of the ash buildup is tin oxide.

We routinely replace our combustion tube every 2000 samples or so.  At this
age there does not appear to be any reason to keep them longer as the cost
per sample by that time is only on the order of US$0.10 per sample for the
combustion tube, and it doesn't seem worth the risk of loosing samples by
trying to hang on to the combustion tube any longer.  However, the
combustion tube frequently seem as though it will go longer than this.

As to safety of either technique I find it hard to believe that the liners
are really all that much safer than sucking out the ash.  I have never
cracked a combustion tube while sucking out the ash.  I would think that
most of the hazard is from the chromium oxide.  We pack our combustion
tubes in a hood, and use a funnel with a long piece of tubing on it to put
in the chromium oxide, so that it doesn't touch the sides.  Even so, there
are traces of Chromium oxide on the sides of the top of the combustion
tube.  Either putting in a quartz liner or suction tube risks contacting
some of this material and spreading it around.  I would have thought that
taking a red hot liner out with would also be dangerous, especially since
it might have chromium oxide on its surface.  Then the ash must still be
disposed of somehow, which in a suction system is taken care of
automatically with little handling.  We do not wash any of our parts as
this means the water from the washing must be collected and disposed of as
chemical waste.  No water that has been in contact with chemicals like
chromium oxide, or even tine oxide, can be poured down our sink.  This
means that washing the tubes out is not an option for us.

We have tried to use quartz liners whenever analyzing glass filter paper,
with mixed results.  This is necessary as it is virtually impossible to
suck out molten glass from the furnace.  We are trying to switch all out
work requiring filter papers to quartz fiber ones that seem to work well
and suck readily out of the furnace.

   With the inserts we would start to get trailing peaks or other problems
that removing the insert seemed to solve.  I have been told by quartz liner
users to either have them long enough so that they sit out of the top of
the combustion tube by a few mm and are pressed in by the sampler, or short
enough so that they don't.

I would certainly like to have the option of using quartz liners.  Is there
any agreement on if they should be rise above the quartz tube and be pushed
down by the sampler, or lower than the sampler?  Should they have slots cut
in them?  How many samples do users get before they need to be
changed?  What flow rate and cycle time for C and N analysis do users with
quartz liners achieve?  For users who do wash their tubes, how do they get
rid of the chemical waste?

Thanks to all who have contributed to this discussion.

Paul.