I found in two sources the following relative ion currents for H2O (the
first column is for 90 eV, for the second column no information was
available):
m/z   %      %
 1   2.4    n.a.
16   1.8    1.1
17  26     23
18 100    100
19   n.a.   0.1
20   n.a.   0.3

However, fractionation patterns depend on some factors like ionisation
energy etc. Recombination seems to be insignificant with H2O, but it might
occur with NO (don't remember the reference, sorry).

With our membrane-inlet quadrupol we have the same problem. For example, the
overlap at m/z 16 makes it impossible to measure CH4 with a membrane inlet
that has a silicone membrane (water permeable). Instead, we measure CH4 at
m/z 15 (relative intensity ca. 80%). Fluorinated membranes might be better,
and I would really be glad to get an idea what to use instead of silicone:
it should have the viscosity of silicone (to be applied to the opening of a
capillary) and discriminate against water, it should harden within some
hours, form a tight bound to steel, be resistant for months with the vacuum
applied .....

Peter
-----------------------------------
Peter Frenzel
MPI fur terrestrische Mikrobiologie
Karl-von-Frisch-Str.
D-35043 Marburg
Tel. +49-(0)6421-178820
FAX  +49-(0)6421-178809, -178823
http://www.uni-marburg.de/mpi/research/frenzel.htm

> -----Original Message-----
> From: Stable Isotope Geochemistry [mailto:[log in to unmask]]On
> Behalf Of Paul Eby
> Sent: Dienstag, 28. Marz 2000 18:48
> To: [log in to unmask]
> Subject: Re: m/z 16 signal
>
>
> >The nature of ionisation would make it extremely unlikely for two ions to
> >'meet' and form a molecule.  The ions are relatively few and far between,
> >at a very low pressure and have the same charge.  I've never seen water
> >dissociating to form mass 16.  This would involve both H atoms being
> >dislodged, so you'd expect to have a huge 18, a smaller 17 (H20
> + e -> HO+
> >+ H) and a smaller still 16 (HO+ + -> O+ + H).
>
> We see exactly this on our Finnigan 252. When water is present
> (mass 18) we
> always have a smaller mass 17 and an even smaller mass 16. Of course, this
> isn't a quadropole....
>
>
>
>
>
>
>
> Paul Eby
>
> [log in to unmask]
> Biogeochemistry Lab
> School of Earth and Ocean Science
> University of Victoria
>
> phone: 250-721-6183
> fax:   250-472-4620