Book Description

The assessment of the potential radiological risk for the human health and the environment that a radioactive contamination episode may entail is one of the most concerning issues in the current society, due to the increasing amounts of radioactive waste resulting from the proliferation of the nuclear industry and other activities. Such assessment is performed by taking into account a large number of environmental components, processes and interactions occurring in a given contamination scenario and governing the transport of radionuclides from the source point of contamination to a potential target individual. In those radioactive contamination episodes in which the terrestrial ecosystem may be affected, one of the key processes to be described, in order to properly assess the potential human exposure to radioactivity, is the partitioning of released radionuclides between water sources and soils in a given contaminated area, since it controls in a great extent radionuclide transport to non-contaminated areas and subsequent introduction into the food chain. Such process can be estimated with the solid-liquid distribution coefficient (Kd), which is a highly operational parameter that for a given radionuclide may remarkably vary depending on the soil-solution system characteristics. In light of this, the aim of this thesis is the development of strategies to provide reliable solid-liquid distribution coefficients (Kd) of radionuclides in soils as input data for models devoted to assess the radiological risk that could be arisen from radioactive contamination episodes involving the terrestrial ecosystem. To reach this goal, two different approaches aiming at reducing and explaining the variability of this parameter and based on the knowledge of the specific radionuclide-soil interaction behaviour, are explored. On the one hand, despite being paramount elements in the field of radioactive waste management and nuclear safety, there is limited knowledge regarding the interaction in soils of trivalent actinides and lanthanides, such as americium (Am) and samarium (Sm), respectively. Due to the scarce and highly variable soil Kd data available in the literature for these elements, and the unclear conclusions about their sorption mechanisms and the soil properties involved, the proposal of reliable Kd data for risk assessment purposes is seriously jeopardised. According to this, the present thesis is focused on the examination of the main factors controlling the interaction of trivalent actinides and lanthanides in soils. On the other hand, due to the lack of site-specific Kd data, radiological risk is frequently assessed with models using, as input data, generic soil Kd values of radionuclides. Such generic Kd data (single values or functions) most of times are statistically derived from compilations in which Kd values may range within several orders of magnitude as a result of contrasting sorption behaviour among the soil-solution systems involved. This fact leads to a high uncertainty in the description or prediction of radionuclide soil-to-solution partitioning in the contamination scenario under assessment. According to this, the present thesis is also committed to the development of a strategy to derive probabilistic Kd data from compilations with low uncertainty and suitable to perform reliable risk assessments.