Howard,

        We routinely perform on-line reductions and D/H-ratio measurements
of small amounts (nmoles) of H2O vapor in such systems and therefore are
particularly sensitive to blanks from moisture.  Our experience is that any
gas extracted from a metal cylinder and any gas exposed to a metal, glass,
or ceramic surface (other than fused silica capillaries) contains water
vapor that is almost impossible to get rid of without aggressive
cold-trapping (presumably chemical getters would also work).  If your goal
is to remove as much water vapor as possible from your analyte stream, I
recommend you put a dry-ice or liquid nitrogen trap on every gas line that
is supposed to carry only non-condensible gas, positioned at the last
possible point before reservoirs containing condensible analytes. You can
reduce the moisture content of gas streams by several orders of magnitude
this way, although even these measures do not completely dry the system
(presumably because the vapor pressure of H2O over ice formed in your traps
is small but measurable).

John Eiler
Division of Geological and Planetary Sciences
Caltech

>Hello all,
>
>I am setting up an IRMS lab and was wonder what experience people have had
>using in line filters on their supply and standards gases. In my past
>experience with GC/MS systems, even using HP and UHP gases (4.5-5.0) it was
>always recommended to have moisture traps on the gases. Has anyone
>experienced problems doing this on a EA-IRMS system? Or do you assume that
>the gases are ultra-clean from the supplier and leave it at that? An
>recommended filters to get or stay away from for IRMS work? Thanks,
>
>Howard
>Howard Sanford
>Stable Isotope Ratio Mass Spectrometry Lab
>919-513-3039
>North Carolina State University
>Dept. of Soil Science
>Box 7619
>Raleigh, NC 27695