Lanthanoid Isotopic Composition of Pre and Post-Detonation Nuclear Material

Abstract

Analysis of lanthanoid isotopic composition of pre and post-nuclear detonation materials provides information on the type of device, origin of fissile material, and in the case of spent nuclear fuel, the operating history of the reactor. Prior to analysis, the lanthanoids must be separated from bulk materials to reduce exposure to harmful radiation and to remove isobaric interferences. Trinitite and spent nuclear fuel rods are appropriate analogues for post and pre-detonation nuclear materials, respectively. Compositional analysis of trinitite glass, fused silicate material produced by the Trinity test, reveal non-normal Nd isotope composition, with deviations of -1.66 ± 0.48 e; (differences in parts in 10<super>4</super>) in <super>142</super>Nd/<super>144</super>Nd, +2.24 ± 0.32 e; in <super>145</super>Nd/<super>144</super>Nd, and +1.00 ± 0.66 e; in <super>148</super>Nd/<super>144</super>Nd (2σ) relative to natural reference materials. Greater isotopic deviations are found in Gd, with enrichments of +4.28 ± 0.72 e; in <super>155</super>Gd/<super>160</super>Gd, +4.19 ± 0.56 e; in <super>156</super>Gd/<super>160</super>Gd, and +3.59 ± 0.37 e; in <super>158</super>Gd/<super>160</super>Gd. The isotopic deviations are consistent with a <super>239</super>Pu based fission device with additional <super>235</super>U fission contribution and a thermal neutron fluence between 0.97 and 1.4 x 10<super>15</super> neutrons/cm<super>2</super>.

Separation and analysis of spent nuclear material is a difficult challenge in both logistics and sample handling. Lanthanoids were removed from the bulk spent nuclear fuel at Savannah River National Laboratories, while the separation of Gd, Sm and Nd was carried out at the University of Maryland. The isotopic composition of Nd and Sm were compared to predicted values calculated using two programs that were developed for modeling the burning cycle of traditional power-reactors: Oak Ridge Isotope GENeration (ORIGEN-S) and Monte Carlo N-particle transport code (MONTEBURNS). The isotopic composition of Nd agreed with predicted values within 10% with the exception of <super>142</super>Nd, while only <super>150</super>Sm had agreement within 10% of prediction. These results show that the typical calculation codes are not adequately modeling the intense neutron flux present in research reactors, and further work will need to be done before source reactors can be identified using reverse modeling algorithms.