Bill,
If there were problems with the rapid precipitation technique, they must
have been worked out because Bill Straight got excellent results with it
here, over a wide range of sizes. I just looked back at some of his
tests and he got std dev's of 0.2 to 0.3 per mil for original sample
sizes of 2 through 10 mg in 1 mg increments. It may not be the technique
of choice below the 2 mg size, Bill would have to address that. Also, I
think you might be referring to a different technique than what he ended
up using later, there were a couple of variants on the rapid precip. The
Dettman et al 2001 method is what he references but again Bill would
have to address the details about any modifications as he did that work.
I wouldn't doubt that every variable in the whole process will affect
results when using BaSO4, but I prefer not to use it based on the
experience here.
For a standard near the value of NBS127, we have gotten excellent
reproducibility with silver phosphate from both Sigma Aldrich and Alfa
Aesar, as well as silver phosphate precipitated from a sodium phosphate
reagent.
I don't think anyone has figured out everything that goes on in the
TCEA, that's the value of these discussions and different labs trying
different approaches.
Jon
Bill Showers wrote:
> Jon,
>
> After Straight's paper came out we ran several runs in the TCEA of
> standards and samples with phosphates, nitrates, sugars and the BaPO4
> stds placed throughout the run (beginning, middle and end), only at
> the end of the run, and then some runs without any BaSO4 stds at all.
> We could not see any difference in the isotopic composition of nitrate
> and phosphate samples run with and without BaSO4 standards. So the
> "scavenging" that you talk about in Straight's paper we could not
> duplicate in our lab.
>
> There was a difference between the way that the phosphates were
> prepared as AgPO4 in our lab and by Straight. AgPO4 prepared by the
> Crowson et al. ion exchange method were fairly consistent isotopically
> across all size ranges as mentioned in our 1991 paper.
>
> AgPO4 samples prepared by the quick precipitation method used by
> Straight were consistent at the larger size ranges, but had a
> significant negative trend below the 5 mg sample size (up to 1.5 per
> mil). These were standards prepared by Straight and run in our lab
> with our calibration routine. At the time we concluded that the
> problem was not with the TCEA, but with differences in the sample
> preparation techniques. The quick precipitation technique should be
> OK with consistent sample sizes between samples and standards, but of
> course this is not always possible with teeth samples that are small
> to begin with. Not like running a good old dinosaur bone with lots of
> sample to analyze.
>
> But the real important issues when running the TCEA is calibrating for
> isotopic non-linearity. To do this you have to use standards that
> bracket the sample isotopic composition. We alluded to that problem in
> our 2002 paper (Showers et al., 2002 Anal. Chem 74(5) 142A-150A). We
> have used NBS-127 because of the low O-18 value along with NIST 8549
> and NIST 8542. That give us a good range from +9.3. to +36.4 per mil,
> but when USGS 34 became available several of us realized that
> something was not right. NBS 127 when run as an unknown was not 9.X
> but ~8.x per mil using USGS 34 as the lower-end isotopic standard.
>
> I notice on the last update from NIST that NBS 127 is now listed as
> +8.6 and not the old +9.3 value, I presume that is the result of the
> availability of USGS 34 (-27.9) and USGS 35 (+57.5) which greatly
> extends the isotopic calibration range thanks to J.K. Bohlke. Using
> NBS 127 with the old value, NIST 8549 and 8542 we were pretty close to
> USGS 35, but not close to USGS 34. This new value for NBS 127
> resolves these calibration issues.
>
> Incidentally, we got some very pure nitrate of soda from a fertilizer
> company in Georgia
> http://www.maconfeedandseed.com/product_info.php/products_id/103 that
> has an oxygen isotopic composition near USGS 35. It was $6.99 for a
> lb, so the shipping cost are much as the fertilizer. You can use that
> and not deplete the USGS 35 for the upper O-18 standard.
>
> But we are still stuck for a uniform and consistent lower end O-18
> standard for pyrolysis. We had thought of taking some depleted water,
> dissolving a bunch of sugar in the water and then freeze-drying it to
> see if the sugar would re-equilibrate and be more negative.
>
> Anyone have any thoughts about a pyrolysis standard near the
> composition of NBS 127 or lower, but without the sulfur?
>
> Best,
> Bill
>
> At 12:17 PM 7/30/2007, you wrote:
>> Bill,
>> the problem with BaSO4 goes away with cleaning the glassy carbon tube
>> and granules in the reactor, so it is only a problem when you run
>> silver phosphate samples immediately after the BaSO4, and the next
>> several samples or standards will be affected. You wouldn't notice
>> unless you ran several silver phosphate standards immediately after
>> the BaSO4, and watch the variation. It does not seem to permanently
>> affect the reactor. The apparent oxygen scavenging effect of the Ba++
>> also occurs with Ca++, another good reason not to run raw samples
>> rather than silver phosphate. We too changed things out frequently,
>> but backgrounds are not the issue, it's reduced yield and
>> fractionation due to oxygen scavenging by the cations.
>> Jon
>>
>>
>> Adversity is an opportunity
>
> William J. Showers
> Dept of Marine, Earth & Atm Sciences
> North Carolina State University
> Raleigh NC 27695
>
> (919) 515 - 7143 Office
> (919) 515 - 7802 Fax
> (919) 515 - 3689 Computer Lab
> (919) 632 - 2283 Cell
> (919) 515 - 7911 Field Lab
>
> Visit our RiverNet Web Site
> http://rivernet.ncsu.edu/
--
Jonathan D. Karr, Ph.D.
Technical Director, Duke EnVironmental stable Isotope Laboratory (DEVIL)
Duke University Dept.of Biology
3322 French Science Center
Science Drive
Durham, NC 27708-0340
(919) 660-7418
Fax: (919) 660-7425
[log in to unmask]
laboratory website: http://www.biology.duke.edu/jackson/devil/
Adversity is an opportunity
|