Textural, compositional, and morphological features of carbonate precipitates were studied and described on hand specimens, slabs, and polished thin sections using light and scanning electron microscopy (CamScan SEM) at the GEOMAR Research Center for Marine Geosciences. The CamScan SEM was equipped with an energy dispersive X-ray spectrographic analyzer (EDAX) that allowed us to obtain qualitative elemental analyses of selected mineral grains.
Bulk mineralogy and the
relative abundance of different carbonate minerals in each sample, were
investigated by X-ray diffraction (XRD) using a Philips PW 1820 X-ray
diffractometer at the GEOMAR Research Center for Marine Geosciences and a
Philips Norelco 1720 X-ray diffractometer at the University of North Carolina at
Chapel Hill, both generating monochromatic Cuk
radiation. Samples were crushed with an agate mortar and pestle, mixed with an
internal standard (
-Al2O3),
and prepared as randomly oriented powder slides. Scans were run from 20º to
60º 2
at a scanning
speed of 0.01º 2
/s. The
relative proportions of different carbonate minerals were quantified on the
basis of the (104) peak areas of calcite, Mg-calcite and dolomite, and the (111)
peak area of aragonite using prepared calibration curves. For details about the
calibration, see Greinert (in press). The position of the (104) peak was used to
determine Mg content of carbonate minerals (Goldsmith et al., 1961; Lumsden,
1979). Calcite with less than 5 mol% MgCO3 is considered low-Mg
calcite (LMC), and all other calcite compositions are referred to as high-Mg
calcite (HMC) after Burton and Walter (1987). Carbon and carbonate contents
(expressed as weight percent [wt%] CaCO3) of both carbonate
precipitates and background sediment samples were determined using a NA-1500
Carlo-Erba element analyzer at the GEOMAR Research Center for Marine Geosciences
and a Coulometrics Model 5011 CO2 Coulometer at the
University of North Carolina at Chapel Hill.
Carbon coated, polished thin sections were made from selected authigenic carbonates and examined by electron microprobe analyses to provide detailed mineral chemistry of carbonate precipitates. All analyses were performed using a Cameca SX 50 electron microprobe at the GEOMAR Research Center for Marine Geosciences, which is equipped with a wavelength dispersive system. During a typical analysis, concentrations of Ca, Mg, Mn, Fe, Sr, Ba, C, O, Al, and P were determined. Counts were collected for 20 s using an accelerating voltage of 15 kV and a beam current of 10 nA in scanning mode (magnification 25,000). Natural carbonate minerals, feldspars, and glasses were used as standards. Matrix corrections were made using standard ZAF techniques.
Samples for oxygen and
carbon isotope analyses were extracted from the surfaces of polished slabs using
a hand-held microdrill. The CO2 for analysis was obtained by reacting
samples with 100% orthophosphoric acid in vacuo at 75ºC (Carbo Kiel CO2
preparation device). The purified CO2 gas was analyzed isotopically
in a Finnigan MAT 252 mass spectrometer at the Geological Institute Erlangen.
Results are expressed in
notation relative to the Peedee belemnite (PDB) standard. Precision is on the
order of 0.1
for both
oxygen and carbon. The
18O
values for dolomite have been corrected following the method proposed by
Rosenbaum and Sheppard (1986). Pore-water oxygen isotope data are from Borowski
et al. (1997).
Stable oxygen and carbon measurements on bulk sediments were measured on a Finnigan Mat 251 Ratio Mass Spectrometer at the North Carolina State Stable Isotope Laboratory. Samples were roasted under vacuum at 325ºC for 1 hr to remove volatile organics and then reacted at 75ºC for 10 min with 100% orthophosphoric acid in side-arm reaction vessels. The CO2 was cryogenically separated for analysis. Because many of these samples contained sulfides, an additional cryogenic separation was done at ~ -130ºC to insure that any SO2 produced during the digestion procedure would be separated from CO2 (Rodriguez et al., Chap. 30, this volume).
The 87Sr/86Sr
composition of 15 carbonate and six pore-water samples was determined using a
VG-Sector 54 Thermal Ionization Mass Spectrometer at the Department of Geology,
University of North Carolina at Chapel Hill. Samples for 87Sr/86Sr
analysis were microdrilled adjacent to samples for 13C
and
13O
determination. Samples were dissolved in 500-µL 1-M ultra-pure acetic acid,
centrifuged, and then the decantate was evaporated dry. Samples were then
redissolved in 250 µL of 5-N HNO3 and loaded into columns containing
Eichrom Sr-specific resin. Details of the procedure are given in Paull et al.
(1995a). Values of 87Sr/86Sr are reported relative to a 87Sr/86Sr
ratio of 0.71025 for SRM-987. Replicate analyses of SRM indicate uncertainties
of ±0.000011 (1
). For
pore-water 87Sr/86Sr analyses, see Paull et al. (1995a).