Re: [asa] Denver RATE Conference (Thousands...Not Billions)_Part 2

Steve,

Thanks for providing the technical details below. Some brief comments:
1) The difficulty in oxidizing the diamond is a concern. I infer
that they resorted to a non-standard recipe (probably higher temp or
longer time) for the oxidation, and they possibly did not get
complete oxidation. Either could affect the results.
2) The use of a very dissimilar material (natural gas) for the
background comparisons is a concern. For a reliable background, one
should use a similar (but radiocarbon-dead) material of similar size
to the unknown, and they should be processed identically. We don't
know if the backgrounds were processed with the same non-standard
oxidation as the diamond, and if the samples were approximately the
same size or not.
3) The reference to a "standard background" is a concern. In
practice, the background will fluctuate with accelerator performance
and is dependent on sample size, the batch of chemicals and glassware
used, the history of the graphitization apparatus, etc. For old
samples like this, the background sample must be processed
identically and in parallel with the unknown and should be switched
back and forth with the unknown in the accelerator to track the
dynamic behavior of the instrument backgrounds.
4) I suspect that two laboratories are involved; an independent lab
which did the sample prep and a leading AMS lab which did the dating
without knowing what the samples were. (I doubt that any of the
leading AMS labs would have knowingly agreed to do these tests for a
YEC group.) If this is so, there is additional room for error due to
the looser coupling between sample prep and AMS. (E.g. The AMS lab
may not have known which samples were supposed to be used as
backgrounds, which is essential as pointed out above.)

I would interpret Baumgardner's diamond results as a background
measurement, mainly of sample prep and instrument backgrounds. The
fact that the results are fairly consistent with one another supports
this. In fact, it's fairly impressive that they can get such a low
background (55k years) for a non-standard sample with non-standard
processing. I suspect the natural gas is lower mainly due to
different processing (different contamination) in the graphitization
procedures.

As you note, the data suggests a difference between alluvial and deep
underground diamond (though it's hard to tell which ones have
backgrounds subtracted, and the difference is similar to the
background). As you suggest, one possibility is modern
contamination. Some other possibilities:
1) different nitrogen contents and neutron activation
2) different nearby radioactive ores
3) cosmic ray activation of the alluvial deposits
4) different sample sizes
5) different graphitization or accelerator backgrounds on different days

Kirk

On Sep 26, 2007, at 8:56 AM, Steven M Smith wrote:

> "Our sample consisted of about 50 mg of sub-millimeter diamond
> chips obtained by shattering a diamond from the Kimberley district
> in South Africa in a sapphire mortar and pestle. Because of the
> laboratory's lack of experience in oxidizing diamond, it required
> several attempts before they were successful. ... The 14C/C value
> for the diamond, which as in the case for the coal samples was a
> composite number based on four separate AMS runs, was 0.096+/-0.026
> pMC, where the precision represents +/-l [standard deviation], or
> 68.3% confidence limits. This number, unlike the coal results
> presented above, does not have the laboratory's standard background
> of 0.077+/-0.005 pMC subtracted from it. The reason for reporting
> the uncorrected measurement here is to be able more clearly to
> compare it directly with the laboratory's standard background
> value. From a statistical standpoint, the result for this diamond
> overlapped, in terms of its confidence limits, the value obtained
> from a much larger number of runs on the purified natural gas the
> laboratory uses as its background standard. The fact that the
> diamond displayed a comparable 14C/C value as the natural gas
> background standard, however, was consistent with our working
> hypothesis that all carbon in the earth contains a detectable and
> reproducible (using the AMS technique) level of 14C." (p. 609-610)
>
> Baumgardner then sent in 6 more diamond samples (including the
> original sample) - 5 from kimberlite pipe mines and 1 from alluvial
> deposits.
>
> "The 14C/C values for these six diamonds cluster tightly about
> the mean value of 0.12+/-0.01 pMC [background not subtracted]. From
> a statistical standpoint, this mean value is consistent with an
> identical 14C/C ratio in all six of the diamonds. With the larger
> number of diamond analyses, there was now a clear statistical
> separation between the mean diamond value and the laboratory's
> background value of 0.08 pMC obtained by repeated runs on a sample
> of purified natural gas. The laboratory concurred with this
> conclusion. We note that by using the usual uniformitarian
> assumptions for converting a 14C/C value into an age (which
> obviously do not apply since these diamonds almost certainly have
> not experienced any recent exchange of their carbon atoms with
> those in the atmosphere), one gets a uniformitarian age from this
> mean 14C/C ratio of 55,700 years." (p. 610-611).
>
> Next Baumgardner sent in an additional 6 diamond samples from
> Namibian alluvial deposits. The results for all 12 diamonds were
> reported in table 6 (p. 614) with the laboratory standard
> background correction applied.
>
> "Although the 14C/C ratios shown in Table 6 are small when the
> standard background is subtracted away, and in some cases are
> smaller than the confidence interval of the measurements from which
> they are derived, they nevertheless are all positive, that is,
> greater than the laboratory's standard background value. If one
> averages the values in Table 6 of the five diamonds from kimberlite
> mines to get better statistics, one obtains the value 0.04 pMC.
> Doing the same for the seven alluvial samples, one obtains 0.12
> pMC. These data suggest, at least from a statistical standpoint,
> that 14C exists in these diamonds to a high degree of certainty, as
> astonishing as that may seem." (p. 613-614)
>
> Although 12 diamonds is not a sufficient number of samples to
> statistically determine average C14 concentration ranges, I find it
> interesting that their own data suggests a possible source for
> excess C14. The kimberlite mine (deep underground) diamonds have
> corrected 14C/C ratios ranging from 0.01-0.07 pMC (average 0.04
> pMC) - the highest values being barely above the confidence
> interval of measurement. The alluvial diamonds (mined from
> surfical stream gravels with presumably 1,000's of years worth of
> exposure to modern C14 levels) have corrected 14C/C ratios ranging
> from 0.03-0.31 pMC (average 0.12 pMC). This suggests to me that,
> despite all the care that may have been taken during sample
> preparation, there is probably still a problem with modern C14
> contamination.
>
> Combine possible modern atmospheric C14 contamination with other
> non-atmospheric sources of C14 (thermal neutrons from U/Th decay
> converting trace N14 to C14; cosmogenic C14; ?) and you *may* have
> an explanation for the slight enrichment in some diamonds. To be
> fair, Baumgardner does address some of these non-atmospheric C14
> sources and concludes that their influence is 4-5 orders of
> magnitude too small to explain the RATE results.
>
> Steve
> (Disclaimer: Opinions expressed herein are my own and are not to be
> attributed to my employer ... or anyone else.)
> _____________
> Steven M. Smith, Geologist, U.S. Geological Survey
> Box 25046, M.S. 973, DFC, Denver, CO 80225
> Office: (303)236-1192, Fax: (303)236-3200
> Email: smsmith@usgs.gov
> -USGS Nat'l Geochem. Database NURE HSSR Web Site-
> http://pubs.usgs.gov/of/1997/ofr-97-0492/

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Thu Sep 27 17:27:45 2007