We have found better performance when a continuous oxygen bleed is used. As
you have found, the oxygen in an oven can be consumed over a day and affect

This can be accomplished using a tank of 1% O2 in UHP He, connected via a
double ferrule between the column and the oven. The required flow is very
low and difficult to set (something like 0.01mL/min), so you need a large
pressure drop between the tank and the oven. A long piece of thin capillary
(0.05mmID for instance) works fine.

You can then set the amount of O2 reaching the oven by adjusting the
pressure at the tank regulator and monitoring mass 32. On the mass 44 cup
(measuring mass 32), we set the oxygen level to about 200mV (of course, you
couldn't do that with a reduction oven after the oxidation oven). With a
constant oxygen flow in, you don't have to worry about reoxidizing ever.

Paul Eby
University of Victoria

At 01:38 PM 5/1/2003 -0400, you wrote:

>Hello everyone,
>         I have run the stable carbon isotope composition of n-alkanes
>standards (20ng/ul;n-C17 to n-C30)in GC-C-IRMS. I am using a Finnigan MAT
>combustion furnace containing Cu, Ni, and Pt wires and maintained at a
>temperature of 940 degree. I have some trouble with this kind of combustion
>reactor since the offset between on-line and off-line was ~-0.5 to -1 permil
>and the negative shift continued to increase gradually after 2 or 3 runs. I
>was wondering if there is anyway to improve the furnace performace.(I
>reoxidized the furnace @940 for ~20' everyday).
>Any input would be greatly appreciated.
>Thanks in advance,
>Wendy Zhang
>Air Quality Research Branch
>Environment Canada
>4905 Dufferin Street
>Toronto, ON  M3H 5T4
>Tel: (416)-739-5792
>Emai: [log in to unmask]
>Incoming mail is certified Virus Free.
>Checked by AVG anti-virus system (
>Version: 6.0.473 / Virus Database: 271 - Release Date: 4/17/2003