Willi, Thanks for the very interesting observation on why the linearity
for CO may vary with trace N2 in the CO reference gas.
On our Delta V we were able to improve the CO 18O linearity by reducing
the emission current to 90 mA, following the recommendation from Todd
Brown at Monsanto. He was good enough to send me data showing the
linearity for 55-126 mA tuning and found 75 to 90 mA was the best linearity.
On 7/11/2012 7:24 AM, Willi A. Brand wrote:
> here is a guess:
> We had different linearities for different CO reference gases. If
> there is a small difference in residual N2 in your CO tank or ref.
> line (I bet there is), the N2 will enter the ion source together with
> the CO. The well known NO production on the filament will produce a
> time-lagged m/z 30 signal, obscuring the pure CO result on mass 30.
> Depending on the width of your reference peaks and on the distance
> between them, you will have a variable cross talk that may look like a
> linearity issue. However, if you vary the signal size by having
> different amounts of material weighed in (and trusting that the
> conversion has no further artifact, which it probably will have), you
> might find out about true linearity issues. We try to get around the
> problem by weighing in equal O-equivalents, so that the
> sample-to-sample comparison is sufficiently well founded in spite of
> the fact that the CO reference might be obscured by traces of N2.
> Sounds reasonable?
> Regards Willi
> On 7/11/2012 16:13, Gregory A Wandless wrote:
>> Anybody got any secrets for improving linearity of 18O? The 13C is
>> fine but 18O in CO (TC/EA) is awful.
Mr. Paul D. Brooks, Spectroscopist,
Dept. Integrative Biology MC3140
1005 Valley Live Sceince Building,
UC Berkeley, CA 94720-3140.
[log in to unmask]