David B. Curtis

Fission product retention in the Oklo natural fission reactors

Eight samples from a planar cross section of one of the natural fission reactor zones at the Oklo U mine were analyzed to determine the abundances and isotopic composition of U, Mo, Ru, Pd, Ag, Cd, Sn, Te and Nd. Fission product concentrations were calculated from the isotopic compositions. The relative abundances of these fission products are different from those produced by fission i.e. portions have been lost or gained from all the samples. The proportions of Te, Ru, 99Ru, Pd and Mo in the eight samples are invariant. We attribute this regularity to retention at the site of production—primary retention. Based upon this interpretation, we suggest that fixed proportions of Ru, Tc, Pd and Mo have been removed from the reactor zone. In contrast, Nd and Sn have been depleted in some regions of the zone and enriched in others. Portions of Ag and virtually all the fissiogenic Cd have been removed from the reactor zone. By analogy with anthropogenic spent fuel, we suggest that the degree of primary retention was controlled by phase assemblages formed in the Oklo natural reactor fuel in response to microscale conditions of pressure, temperature and composition produced by the nuclear reactions. The ability of these putative minerals to retain nuclear products was a function of their stability under physiochemical conditions established by the geological environment over the last 2 Ga. An integrated study of the natural reactors would test our hypothesis and provide valuable information for evaluating the long-term effectiveness of anthropogenic spent fuel as a container of nuclear waste in the geochemical environment.

A revision of the meteorite based cosmic abundance of boron

Abstract Analyses of meteorites for B abundances have shown that many chondrites are contaminated with terrestrial B, producing erroneously high meteoritic abundances of this element. Boron concentrations in freshly prepared interior samples are significantly lower than they are in samples with unknown or unspecified terrestrial histories. An estimate of the cosmic abundance based upon the analyses of 8 interior samples of 2 carbonaceous chondrites and 1 interior sample of each of 8 ordinary chondrites is a factor of 6.7 less than the previous low estimate. Our revised value, 3.0 B/1010H, is in excellent agreement with estimates based on observations of the solar photosphere. There is no longer a need to consider processes that enrich B in carbonaceous chondrites or deplete it in the sun. Relative meteoritic abundances of Li, Be and B are now in general agreement with models of nucleosynthesis of these light elements by galactic cosmic ray induced spallation.

Analysis of naturally produced technetium and plutonium in geologic materials

In uncontaminated natural materials, plutonium and technetium exist exclusively as products (daughters) of nuclear reactions in which uranium is the principal reactant (parent). Under conditions of chemical stability over geologic periods of time, the relative abundances of daughter and parent elements are fixed by the rates of nuclear reactions and the decay of the daughter radionuclide. The state of this nuclear secular equilibrium condition is the primary basis of the geochemical study of these elements in nature. Thus, it is critical that nuclear parent and daughter abundances are measured in the same sample. We have developed a quantitative procedure for measuring subpicogram quantities of plutonium and technetium in gram quantities of geologic matrices such as uranium ores. The procedure takes advantage of the aggressive properties of sodium peroxide/hydroxide fusion to ensure complete dissolution and homogenization of complex materials, the precision provided by isotope dilution techniques, and the extreme sensitivity offered by thermal ionization mass spectrometry. Using this technique, a quantitative aliquot can be removed for uranium analysis by isotope dilution thermal ionization mass spectrometry or α spectrometry. Although the application of the procedure is unique, the analytical concepts may find more general application in studies of environmental contamination by nuclear materials. To assess the precision and accuracy of the analytical results, blanks and standards were analyzed routinely for a 1-year period to ensure quality control of our sample analyses. The average technetium blank is 5 ± 4 fg (n = 8), and that for plutonium is 0.17 ± 0.15 pg (n = 7). Thus, the detection limit for technetium (defined as 3 times the standard deviation of the average blank) is 11 fg, and that for plutonium is 0.44 pg. To assess the procedural precision, Canadian Reference Material BL-5 was analyzed routinely with samples. The results of seven replicate analyses for technetium in this standard reference material yield a technetium concentration of 59.0 fg/g, with a remarkably small standard deviation of 0.6 fg, 1.0% of the average value. The results of six replicate analyses for the concentration of plutonium in BL-5 give 1.012 pg/g, with an equally small standard deviation of 0.016, 1.6% of the average value. No direct measure of accuracy can be done on the technetium or plutonium analyses, because no standard reference material exists for these elements. To help constrain the accuracy of our measurements, equilibrium technetium/uranium and plutonium/uranium abundances were calculated using the nuclear reaction code MCNP. For technetium, such calculations are relatively insensitive to variations in model parameters, and measurements fall within a 21% high/low bias. For plutonium, the calculations are very sensitive to model parameters and hence inherently less precise. Indirectly, spike and isotope mix calibrations made from weighted quantities of certified isotopes (both technetium and plutonium) can be used to determine the bias of the measurement system for these elements. These calibrations show that the measurement system is biased by no more than ±1.5%.

NATURE'S UNCOMMON ELEMENTS : PLUTONIUM AND TECHNETIUM

Abstract Natural 99Tc and 239Pu were measured in ores from the Cigar Lake uranium deposit, in which U concentrations ranged from 0.3 to 55 wt%. Atomic ratios ranged from 1.4 × 10−12 to 51 × 10−12 for 99Tc/U and 2.4 × 10−12 to 44 × 10−12 for 239Pu/U. Measured concentrations are compared to those expected if the ores had behaved as closed systems with respect to U and its products. Under conditions of secular equilibrium in closed systems, 99Tc and 239Pu concentrations are solely a function of the neutron flux, U content, and decay rates. The neutron production rate and physical and chemical parameters that control the in situ neutron flux were measured in several samples. Neutron transport modeling of the sample environs using the MCNP code indicated that about half of the samples showed apparent excesses of 99Tc and 239Pu beyond the amounts predicted for secular equilibrium. Although production by neutron-capture on 98Mo complicates the accuracy of the 99Tc predictions, the excess quantities nonetheless strongly suggest the redistribution of these elements within the deposit. The failure to observe complementary deficiencies within the deposit suggests that the redistribution processes enriched the elements, removing small proportions of the elements from large uraniferous masses of rock, and concentrating them in smaller, less uraniferous, masses. The consistency of 239Pu/U and 99Tc/U ratios in bulk rock suggests that the redistribution processes observed at Cigar Lake are highly localized and in no case result in large-scale losses or gains of these nuclear products from the deposit as a whole.

Fission-product retentivity in peripheral rocks at the Oklo natural fission reactors, Gabon

Abstract Thermal ionisation mass spectrometry has been used to measure the isotopic and elemental abundances of Mo, Ru, Pd, Ag, Cd, Sn and Te in a number of sandstones and shales in the immediate vicinity of Reactor Zone 9 at the Oklo mine site. The mass spectrometric isotope dilution technique was used to measure the elemental abundances of these elements. The data show that considerable amounts of the fissiogenic Mo, Ru, Pd, Ag and Te which escaped from the reactor zone, have been retained in the peripheral rocks, although Cd and Sn were not contained to any significant extent in these samples. Some information on the retention of Sb and Tc was also obtained. Evidence of element fractionation has also been demonstrated in that 99 Tc and Sn were fractionated from Ru and Te, respectively within 1 Ma of the end of reactor criticality. The presence of excess 125 Te in some of the peripheral rock samples indicates that Sb was mobilised from Reactor Zone 9 within tens of years of the completion of reactor criticality. The implications of these results to the storage of radioactive wastes in natural geological repositories is discussed.

Multielement analysis of major and minor elements by thermal neutron induced capture gamma-ray spectrometry

The utility of prompt gamma-rays from thermal neutron capture for the measurement of nine elements (Si, Al, Fe, Na, K, Ca, Ti, Mg and P) in major and minor abundance has been investigated. National Bureau of Standards (NBS) and United States Geological Survey (USGS) standards were used to demonstrate generally good agreement between experimental measurements and certified values. Usually accuracies and precisions of ±10% were observed.

Geochemical controls on 99Tc transport and retention

Abstract Measurements of the isotopic abundance of 99 Ru in rocks containing fossils of natural nuclear fission reactors show that 99 Tc was geochemically active during nuclear criticality 2 Gyr. ago. The element is deficient in the reactor zones and enriched in the rocks peripheral to the zones of criticality. Theoretical consideration suggests that changes in the chemical environment produced by increased temperatures and/or high fluxes of radioactive emanations may have increased the elements solubility in a coexisting aqueous phase. The increased solubility allowed portions of the fission product to be transported away from the reactor zones by the mobile fluid.

99Tc, Pb and Ru Migration Around the Oklo Natural Fission Reactors

The generally favored method for disposal of commercially generated radioactive wastes is deep geologic burial (1). Existing information supports the belief that appropriate geologic media in conjunction with engineered barriers will be adequate to isolate the waste from the biosphere (1). Because of the long half-lives of many of the radionuclides, the required periods of isolation are much longer than recorded human history, therefore evaluation of the reliability of particular geologic media to isolate the waste must rely heavily upon analytic models of radionuclide transport in geologic environments. Studies of element migration in natural systems can yield important constraints for such analytic transport models. Of particular value are the remains of naturally occurring fission reactors found at the Oklo uranium mine in Gabon, Africa. The uranium rich zones began to undergo sustained nuclear fission 2×109 years ago and remained critical for several hundred thousand yqs (2). Approximately 10 tons of fission products and 4 tons of Pu were generated. Hence, Oklo provides a unique opportunity to study movement of fission-produced materials which have resided in a geologic media for times greater than those of concern in the storage of radioactive wastes (3,4). Such studies will be valuable in evaluating transport models and can provide information regarding chemical and physical conditions, time scales, and processes operative in the movement of elements from a well-characterized reactor waste source.

Simultaneous determination of nitrogen, carbon, and hydrogen by thermal neutron prompt γ-ray spectrometry

Abstract Thermal neutron prompt γ-ray spectrometry is applied to the determination of nitrogen, carbon, and hydrogen in environmental materials. Useful, simultaneous, nondestructive analyses with 1-g samples can be achieved at concentrations greater than 500 ppm, 10% and 5 ppm, respectively.

Half-life of /sup 148/Gd

Gadolinium isotopes were separated from a tantalum target which had been irradiated with 800-MeV protons. The concentrations of /sup 148/Gd in weighed aliquots of a solution of the isotopes were determined by isotope dilution mass spectrometry. Other aliquots were counted on a low geometry alpha counter to determine the disintegration rate of /sup 148/Gd per unit weight of soulution. From the data obtained the half-life of the isotope was found to be 74.6 +- 3.0 yr.