HI Jordon--

You'll need a better GC column, as Ben said.

If you have a large N2 peak, dilute it or, if possible, cut it out.  Why?  As Libby mentioned, you can form NOx's in the source--and they tend to take time to pump out.  That gives you m/z 30 interference that varies with N2 signal and a  m/z 30 baseline that changes over time.  When I tried nitrates on the TC/EA (different system, same principle), I needed to dilute the N2 peak to get reliable readings off of CO. 

take care,


On 7/2/2019 10:52 AM, Munizzi, Jordon wrote:
[log in to unmask]">


Sorry in advance if I’m being dense, but how would you exclude the N2 from the IRMS (using 100% sample dilution) if N2 and CO are co-eluting from the EA? I understand the issue of mass 30 interference from NO, but to me (with all of two years of real-world experience under my belt J) this seems like a predominantly chromatographic issue.

Reporting live from peak 1 of the Dunning-Kruger curve,



Jordon Munizzi, PhD

Research Facility Manager

Department of Earth and Environmental Sciences

University of Kentucky

Ph: (859) 270-9177

Fax: (859) 323-1938

[log in to unmask]


From: Stable Isotope Geochemistry <[log in to unmask]> on behalf of Libby Stern <[log in to unmask]>
Sent: Tuesday, July 2, 2019 10:56:46 AM
To: [log in to unmask]
Subject: Re: [ISOGEOCHEM] N2 / CO separation
Hello Dmitry,

There are a number of paper which discuss this issue (Werner et al. 1996, Farquar et al. 1997, Werner and Brand 2001; Bohlke et al. 2003; Gehre and Stauch 2003; Accoe et al. 2008, Brand et al. 2009; Qi et al. 2011, Hunsinger and Stern 2012).

The N2, forms NO in the ion source and creates a persistent high mass 30 background, adversely affecting d18O measurements.  You have a Conflo IV, which permits 100% dilution/exclusion of the N2 from the ion source, so you should add that 100 % dilution to you method. The addition of a longer column or lower temperature can improve the separation, but that will just improve the post-dilution baseline.

Libby Stern

On Tue, Jul 2, 2019 at 9:57 AM Dmitry Kopylov <[log in to unmask]> wrote:
Dear all,

I'm trying to arrange the measurement of d18O in N-containing organics (in hair).
The problem is this substance gives a large N2 peak which is registered on the same collectors with CO. The N2 peak stays so close to CO peak that it leaves no time for correct background detection and leads to big errors in d18O measurement.
I have tried to decrease the reactor and column temperatures, to increase the start slope / end slope settings, to burn the column at 190 C for several hours. It helps a little, but doesn't solve the problem completely.
Does anybody know the ways to improve the separation of these two peaks?
Maybe there is a possibility to use another type of column? In the Thermo catalogue there is only one type of column for d18O analysis, but possibly there are other options?
I'm attaching the example of spectrum acquired with method: 1450 / 60 (Furnace / Oven temp.), 100 / 100 (Carrier and Reference flows).
My equipment is DeltaV Advantage + ConFlo IV + EA Isolink Flash IRMS.

Will be wery grateful for your advices!

Best wishes,
Dmitry Kopylov


Libby Stern