Dear Kinga, Pier, and others interested:

I seem to recall discussions some time ago suggesting that during
on-line analyses of oxygen isotope ratios of silver nitrate, NO may form
in the ion source resulting in somewhat elevated mass 30 signals and
hence leading to high apparent d18O values. It was suggested that the
extent of this effect is likely dependent on filament material, and
hence instrument manufacturer. Could this be another reason why oxygen
isotope ratio measurements on nitrate by on-line techniques provide
typically higher values (e.g. ~25 per mil for IAEA NO-3) than off-line
determinations (~23 per mil IAEA NO-3)? Does anybody have new insights
on this?


Bernhard Mayer
Applied Geochemistry Group (ES 506)
Department of Geology & Geophysics
University of Calgary
2500 University Drive NW
Calgary, Alberta, Canada T2N 1N4
phone: (403) 220 5389
fax: (403) 220 7773
email: [log in to unmask]

-----Original Message-----
From: Stable Isotope Geochemistry [mailto:[log in to unmask]] On
Behalf Of Pier de Groot
Sent: Thursday, December 11, 2003 7:58 AM
To: [log in to unmask]
Subject: Re: d 18O value for IAEA-NO-3

Hi Kinga,

Good you add 'some pieces of the puzzle'. I thought that the value I
gave in
my first message was 'just the best one'. Naive, of course, to think
were that simple.... I had no time to look into the reason of the
difference, so it is good you informed us.

Now there are two reasons my given value is a good one:
1) the reason given by Kinga Revesz (no exchange with a glass wall at
elevated temperature in the on-line method against exchange in the
2) included cap-delta17O measurement for proper correction on 18O

To avoid exchange of oxygen with glass walls, as is the problem in the
off-line methods as mentioned by Kinga, did anybody try out the Ni-tube
method as is used for oxygen analysis of O-containing organic compounds
water? This probably will be done the best way by adding 'carbon powder'
KNO3. I am not sure, but Hg of the Hg(CN)2 might form an amalgam with
the Ni
and therefore is not a good idea (?).
Or do I overlook something here? Would be interesting to see the value


> "It also must be notified that there is a clear difference between
> 'off-line'
> (CO2, O2) and 'on-line' (CO) methods of roughly 2 - 3 per mil.,
> the
> reason. Worrying enough...."
> There is an explanation for that difference:
> The off line methods are using high temperature (520 or 800 C), sealed
> tube, carbon combustion  to produce CO2 out of KNO3. At  that
> there is an isotope exchange between the oxygen of the CO2 and the
> of the combustion tube (quartz, Vycor, Pyrex).   Therefore, the
> ratio scale factors of the "off line" methods are compressed to 0.3 -
> that of the scale factor of the on line method.   The oxygen isotopic
> of the quartz is about 17 per mill, consequently if the oxygen isotope
> the nitrate around +23 permill (such as it is of the  IAEA N-3), the
> difference between on and off line method would be about -(2-3) per
> However, if the O-18 in nitrate is about +54 per mill, the difference
> be about -10 per mill, and if the oxygen of the nitrate is -30 per
> the difference would be around +15 per mill.
> This is demonstrated in our paper "Comparison of delta O-18
measurements in
> nitrate by different combustion techniques.  by Revesz and Boelke,
> Chem., 2002, 74, 5410-5413.
> Cheers:
> Kinga
> Kinga Revesz
> Chemist, Stable Isotope Laboratory
> U.S.Geological Survey
> MS 431
> Reston, VA 20192
> T:703-648-5865
> FAX: 703-648-5274
>                     Pier de Groot
>                     <        To:
[log in to unmask]
>                     NDORA.BE>                cc:
>                     Sent by: Stable          Subject:  Re: d 18O
value for
>                     Isotope
>                     Geochemistry
>                     <[log in to unmask]
>                     UVM.EDU>
>                     12/09/2003 09:26
>                     AM
>                     Please respond to
>                     Stable Isotope
>                     Geochemistry
> Moritz and Eddy,
> The value I actually gave was the value as given in one of my book
> on standards by Manfred Groening, referring to the Boehlke et al.
> paper. I looked again in the Boehkle et al (2003) paper, and indeed, a
> number of values is given.
> I am not certain why Manfred selected this particular value (can you
> yourself on this, Manfred?), but I guess since Boehlke and coworkers
> into consideration the 'capital delta'17O signature (quoted is the
work by
> Greg Michailski and coworkers: Michailski et al. (2002) Anal. Chem.,
> 4989-4993) of IAEA-NO-3 (= IAEA-N3) in their determination of the d18O
> value, it was considered the most reliable determination of the d18O
> that reason!
> It also must be notified that there is a clear difference between
> 'off-line'
> (CO2, O2) and 'on-line' (CO) methods of roughly 2 - 3 per mil.,
> the
> reason. Worrying enough....
> In any case, we really need to have consensus about the values of
> and reference materials. Having a list of values for the same standard
> reference material is not serving anybody.
> If the given d18O value is the best one (I followed Manfred Groening
> this) might raise dispute by some. But we have to decide which value
we do
> accept! We need to have an isotopic value solidly calibrated and
> to the N-isotope scale - that should be the choice. Otherwise we
> compare with each other and may bias our interpretations.
> Pier.
>> Eddy,
>> The Boehlke paper is indeed the latest report on d18O of IAEA-N3. But
>> aware that there are quite a number of other (lower) values reported
>> parts compiled in the Boehlke paper).
>> Moritz
>> On Tuesday, December 9, 2003, at 05:16  AM, eminet wrote:
>>> Dear list members,
>>> Has anyone of you a d18O value for the reference IAEA-NO-3 (for
>>> d15N=4.7)?
>>> Thanks,
>>> Eddy
>>> Eddy Minet
>>> Centre for the Environment
>>> Trinity College Dublin
>>> Dublin 2 (Ireland)
>>> e-mail: [log in to unmask]
>>> tel: +353(0)1 608 2403
>>> mobile: +353(0)86 827 4921
>> Moritz F. Lehmann
>> Department of Geosciences
>> Princeton University
>> Guyot Hall
>> Princeton, NJ 08544
>> phone: (609) 258-7544
>> fax: (609) 258-0796         e-mail: [log in to unmask]