We have been running this method for some time. Out gas bench GC has the
same limitations. This doesn't effect our results. We bake it out, on
occasion, at maximum (~220*C). Its mostly water you want to remove as
that will effect the separation of N2 and O2. If you see good separation
and, your column is probably fine.
Do you scrub volatile organic compounds prior to N2O pyrolysis? This is
critical. If any VOCs enter the hot gold tube with your N2O, the oxygen is
bonded and won't form O2. If you don't scrub, test by dipping some of your
tubing into a -110*C bath immediately upstream of the gold tube. If you
have not been scrubbing VOCs, it might take a while to get them all off
the side walls of the tubing. I suggest heating the tubing while helium is
on with a heat gun for a while. We also periodically purge the gold tube
with pure O2 while its hot overnight. Does your nitrogen signal remain the
same size while the oxygen signal degrades? If so, its VOCs.
Earth and Space Sciences
University of Washington
Seattle, WA 98195
[log in to unmask]
On Thu, 26 Nov 2009, Samuel David James Russell wrote:
> Hello All,
> I am using a 30 m Restek RT-Msieve 5A PLOT column for separations of
> nitrogen and oxygen on our GasBench II. Since October I have been
> attempting to isolate the cause of diminished oxygen signal. I had
> initially excluded the column as the cause because I had baked it out (300
> deg. C for three hours) in an attempt to resolve the problem (as is
> suggested by the manufacturer). During normal runs (thermal cracking of
> nitrous oxide into oxygen and nitrogen) I have observed the oxygen signal
> progressively drop off with every subsequent sample (the first sample would
> have ~3000 mV signal and by the third sample - an hour later - the signal
> would be 200 mV on mass 34). With a rest period the pattern would start
> from the beginning.
> I have injected Argon, oxygen and nitrogen right before and after the
> column. Argon is fine before and after the column but oxygen has a signal
> of ~3000 mV (m=34) if injected downstream of the column and would initally
> look good upsteam of the column but would eventually drop to 200 to 500 mV.
> Nitrogen was not initially tested because it had been fine all along. I
> have baked the column a second time (but for 6 hours) and the problem
> appears to have become worse. Now both oxygen and nitrogen have a constant
> low signal (<500 mV when they should be >3000 mV) when injected upstream of
> the column, Argon is still fine.
> I should note that since our GasBench GC box will not heat up to 300 deg.C
> (see my other post) I have been baking out the column on our GCC and quickly
> moving it over to the Gasbench. I have done this previously with no oxygen
> and nitrogen signal problems.
> It has been suggested to me by our supplier that there could be a problem
> with the helium scrubber which is getting a little old (we only use 5.0
> grade helium). We haven't observed any issues with our EA, TCEA, other
> GasBench or GCC.
> Does anybody have a suggestion? Is our 6 month old column dead?
> Thank you!
> Sam Russell
> Ph.D. candidate
> The University of Western Ontario