Radiochemical Analyses of U, Th Isotopes, Pa and Be

Robert F. Anderson

General

We will measure vertical profiles of naturally-occurring uranium-series radionuclides (U, Th and Pa) as well as cosmogenic Be in dissolved and particulate phases throughout the water column at the location of each EqPac sediment trap mooring. Annual composites of material collected by each sediment trap will be analyzed, as will be fine-grained particles collected by in situ filtration. Individual time-series samples will be analyzed from the site showing the greatest seasonal variability of particle flux. Detailed profiles of radionuclides will be measured in the uppermost 10 to 20 cm of sediments collected by box cores, and additional samples from dated piston cores will be analyzed, going back to ~200 ky BP.

Collaboration

Analysis of both filtered and trapped particles will be coordinated with M. P. Bacon and J. K. Cochran to avoid unnecessary redundancies.

Water Samples

Six 50-liter water samples will be collected near the location of each sediment trap mooring from a depth of ~800 m to the bottom by combining water from paired 30-liter go-flo bottles on a 12-bottle rosette. Unfiltered water is processed to obtain total radionuclide activity. Water is transferred to a 100-liter polyethylene tank and acidified to a pH of ~1.5 with redistilled HCl. Isotope yield monitors ( Th, Pa), Be carrier, and~1000 mg of iron as FeCl (cleaned by extraction into isopropyl ether and back extraction into water) are added to acidified samples and allowed to sit for ~24 hours. Ammonia is then added to raise the pH to between 7 and 8 to induce precipitation of FeO, which carries with it the radionuclides of interest. Precipitates are concentrated by settling. Supernate solutions are decanted and discarded. Precipitates are shipped back to the lab for purification of the radionuclides.

Iron-hydroxide precipitates are freed of residual salts by resuspension in high purity water, and further concentrated by centrifugation. Precipitates are dissolved in redistilled HCl. Fe is extracted into isopropyl ether and discarded. The acid phase, containing the radionuclides, is evaporated by heating, with addition of HNO3 and HF. Final purification of the radionuclides is by ion exchange following a scaled-down version of the procedure described by Anderson et al. (1990). Th and Pa are measured by mass spectrometry. 10Be is measured by accelerator mass spectrometry (AMS). Uranium is not measured because its concentration in seawater can be calculated from the constant U/salinity ratio.

Particulate Matter and Sediments

Dried and weighed samples of particulate matter are totally dissolved in the presence of U, Th and Pa isotopic yield monitors and Be carrier using a mixture of HCl, HNO , HF and HClO following the method of Fleisher and Anderson (1991). Individual nuclides are purified by ion exchange, following a scaled-down version of the procedure described by Anderson et al. (1990). U, Th and Pa are measured by mass spectrometry; Be by AMS. Analysis of sediments follows the procedure of Anderson et al. (1990), including total dissolution of sediments and purification of nuclides by ion exchange. Be will be measured by AMS. At the time of this writing, it is uncertain whether U, Th and Pa analyses will be by alpha spectrometry or by ICP-MS.

QA/QC

All analyses will follow isotope dilution procedures, which eliminates uncertainties associated with unknown or variable chemical yield. Procedural blanks will be evaluated by carrying a small volume of high purity water through all feasible steps of the procedure; collection of water using go-flo bottles is the exception. Our shallowest sample at each station is positioned to coincide with the deepest sample collected using an adsorber cartridge attached to J. K. B. Bishop's large-volume in situ filtration system. Cartridge samples will be analyzed by T.-L. Ku's group for the same long-lived radionuclides. This overlap of sampling will both provide an opportunity for intercalibration and link our deep-water profiles with Ku's detailed profiles of the upper water column. Aliquots of a homogenized sediment sample will be distributed for intercalibration of methods for solid phase analyses (i.e., both particles and sediments, since the methods are essentially the same).

Literature Cited

Anderson, R. F., Y. Lao, W. S. Broecker, S. E. Trumbore, H. J. Hofmann and W. Wolfli (1990).
Boundary scavenging in the Pacific Ocean: A comparison of Be and Pa. Earth and Planetary Science Letters, 90: 287--304.

Fleisher, M. Q. and R. Anderson (1991).
Particulate matter digestion (from mg to 10's of g) and radionuclide blanks. pp. 221--222, In: D. C.
Hurd and D. W. Spencer (eds.),
Marine Particles: Analysis and Characterization, Geophysical Monograph 63, Americal Geophysical Union, Washington, D. C.