Dear Reed and others,
Over the last few years I've performed analyses on several different types
of modern biogenic carbonate and have conducted tests on non-biogenic
calcite and aragonite. I've also read a lot of papers which have discussed,
in passing, methods designed to remove organic contamination prior to reaction
with phosphoric acid. My investigations suggest that the problem lies only with
modern biogenic carbonates - ancient carbonates seem to be largely impervious
to different pretreatments (although as Max Coleman has just pointed out
samples with high organic carbon contents may also be problematical) - and
that 2 sets of published isotopic data for modern biogenic carbonates exist:
1) Thermally treated data (usally heated under vacuum for 1 hour at 380oC).
2) Data obtained following plasma ashing or use of a liquid bleaching agent.
The data obtained from these 2 techniques are usally different for the same
starting material by up to 1 per mil (when authors report comparative studies).
As a result I've drawn what might be considered a controversial conclusion-
about half of the published data on modern biogenic carbonates is erroneous:
i.e. data obtained using one or other of these pretreatment techniques can't
both be right except where the sample isotopic values are obtained for tests
run on material using these procedures.
At the risk of boring everyone to death I'd like to briefly report my
experiences with brachiopods, bryozoans and otoliths. All samples were collected
alive and therefore can be considered as fresh biogenic carbonates. As
everyone is probably aware brachiopods precipitate low Mg calcite, bryozoans
low to intermediate Mg calcite and fish aragonite. The treatments I tried
were as follows:
1) Heated under dynamic vacuum at 380oC for 1 hour except otolith carbonate
which was heated at 200oC (tests show that aragonite progressively inverts to
calcite at higher temperatures when heated for 1 hour; complete inversion
occurs at about 440oC when heated for this duration). Note that the literature
is perculiar in that it seems that it's OK to heat calcite to 380oC, but
aragonite should be heated to 200oC. If 200oC treatment is OK to remove
organic matter from aragonite sample why is it necessary to use a higher
temperature for calcite? Is organic matter in calcitic biogenic carbonate
more thermally resistant than organics in aragonitic shells. Thermal treatments
really need to be tested across a broad range of temperatures- say 100oC to
400oC.
2) Plasma ashed for 2, 4 and 6 hours.
3) Digested for 24hrs in 5% sodium hypochlorite buffered to pH 9 (brachiopod
and otolith carbonate only).
4) Untreated. (A warning here the organics in some modern biogenic carbonates
- notably gastropods - are known to cause severe contamination of mass
specs if not removed. Be extremely careful- prepare some CO2 and run it
through a quadrupole to identify any problem contaminants).
Following routine analysis I found that sub-samples of brachiopod calcite gave
the same isotopic compositions for both carbon and oxygen when run untreated,
plasma ashed and bleached, but both carbon and oxygen values were about 1 per
mil more negative when the same material was vacuum roasted. Bryozoans mostly
gave the same values for untreated and plasma ashed sub-samples, but this
time, while the carbon values obtained from roasted samples were similar to
the untreated and plasma ashed samples, oxygen isotope compositions were more
negative by around 1 per mil. Otoliths gave the same isotopic values
regardless of treatment. 3 different sources of biogenic carbonate, 3
different responses to pretreatment.
I happen to have the bryozoan data to hand so I list it below for examination.
Bryozoan species Pretreatment No. of analyses d13C (V-PDB)d18O(V-PDB)
Hornera robusta
> Untreated 5 +1.25 (0.018) +1.37 (0.047)
>
> Oxygen plasma furnace (2 hours) 5 +1.27 (0.022) +1.32 (0.054)
>
> Oxygen plasma furnace (4 hours) 5 +1.24 (0.024) +1.25 (0.061)
>
> Oxygen plasma furnace (6 hours) 4 +1.28 (0.013) +1.28 (0.022)
>
> Vacuum roasted (375oC, 1 hour) 10 +1.23 (0.039) +0.47 (0.062)
>
> Celleporina grandis
> Untreated 5 +0.48 (0.079) +0.91 (0.102)
>
> Oxygen plasma furnace (4 hours) 5 +0.66 (0.014) +1.05 (0.099)
>
> Vacuum roasted (375oC, 1 hour) 5 +0.60 (0.039) +0.03 (0.038)
>
>
> Myriapora truncata
> Untreated 5 -0.51 (0.032) +1.16 (0.026)
>
> Oxygen plasma furnace (4 hours) 5 -0.51 (0.019) +1.23 (0.059)
>
> Vacuum roasted (375oC, 1 hour) 4 -0.57 (0.007) +0.48 (0.007)
>
> Figures in parentheses are 1 standard deviation.
>
>
I make no bones about the fact that I don't understand the reasons for the
differences, although I have some ideas. Clearly, there is a need to
investigate this problem extremely carefully, but who has the time and money
to carry it out and which grant awarding body would support such a research
project. Of course anyone trying to work on this problem faces a catch 22
because it strike me that it is impossible to obtain test biogenic carbonate
which is not already contaminated in some way. How can one assess the cause
of these isotopic changes? Some workers have, I think, tried to spike
non-biogenic carbonate with various substances, but this is a poor approach
because part of the problem may arise from the lower thermodynamic stability
of biogenic carbonates when thermally treated as opposed to original ideas
which suggest that isotopic exchange might take place between carbonate and
organic matter when samples are heated to high temperatures. What affect does
the presence of fluid inclusions in biogenic carbonate have?
After all this one might ask which set of data are right. I've been using
data to try to assess whether particular taxa precipitate their skeletal
carbonate in so-called isotopic equilibrium by comparison with the
temperature-dependent equation of Friedman & O'Neil. As one can see there
could be a temperature difference of around 4oC for bryozoans depending
on which pretreatment is used. Of course one has to decide which is the
most acceptable data and, on the basis of prejudice alone, I have decided
that the least aggressive technique of plasma ashing is OK. Of course
I'm willing to accept that I'm wrong providing that everyone else who
also uses these low temperature methods also withdraw all of their data.
Likewise, if I'm right then all the published data based on roasting should
be considered suspect unless the authors can demonstrate otherwise.
As far as I can see this will remain a considerable problem to the detailed
interpretation of stable isotope data in modern skeletal carbonate until
the problem has been examined for a wide range of pretreatments and taxa.
Any volunteers,
Steve Crowley
Stable Isotope Lab.
University of Liverpool.
UK.
|