Hi Dachun,


I had a look through all my 18O (and 2H) data from water samples since
we had our TC/EA system installed just over a year ago, and I have never
seen a "memory" for 18O of the extend you describe.

> SLAP: -29.8, -41.6, -46.7, -46.6, -48.7, -48.5, -48.7
> Change syringe.
> Sample: -20.7, -15.8, -14.3, -13.4
> Expected value of sample is about -10.7

Not even our 2H values were that far out (max. difference was 7 o/oo).

That said, we observed some "dodgy" values at the very beginning when we
used a glassy-C tube filled according the manufacturers instructions,
which were optimised for analyses of solid samples.

For analysis of liquid samples, we fill our reactor tube differently to
minimise the dead volume. In other words, we fill the glassy-C tube with
crushed glassy-C chips just shy of the graphite insert.

I don't know if this is the solution to your problem but hope it will
help.


Cheers,

Wolfram



> -----Original Message-----
> From: Stable Isotope Geochemistry
> [mailto:[log in to unmask]] On Behalf Of Dachun
> Sent: 02 April 2004 18:22
> To: [log in to unmask]
> Subject: d18O of water
>
>
> Hi,
>
> My working standard of water for d18O has value of -11.6
> calibrated against SMOW and SLAP using CO2 equilibration. I
> cannot get the same calibrated value when I use glassy-C
> method. The value is about 0.5 heavier. Can anybody help me
> understand this.
>
> The only thing I worried is the carry over effect. I found
> the memory can carry over up to next 6 analysis from SLAP to
> SMOW. Even I used the seperate syringes.
>
>
> Cheers.
>
> Dachun Zhang
> Isotope Lab
> Zymax
>