Realistic bandwidth estimation in the theoretically predicted radionuclide inventory of PWR-UO2 spent fuel derived from reactor design and operating data

Nuclear energy for power generation produces heat-generating high- and intermediate level radioactive waste (HLW and ILW) for which a safe solution for the handling and disposal has to be found. Currently, many European countries consider the final disposal of HLW and ILW in deep geological formations as the most preferable option. In Germany the main stream of HLW and ILW include spent fuel assemblies from nuclear power plants (NPPs), the vitrified waste and compacted metallic waste of the fuel assembly structural parts originate from reprocessing plants. An important task that occurs within the framework of the Product Quality Control (PQC) of nuclear waste is the assessment of the compliance of any reprocessed waste product inventory with the prescribed limits for each relevant radionuclide (RN). The PQC task is to verify the required quality and safety of nuclear waste prior to transportation to a German repository and to avert the disposal of non-conform waste packages. The verification is usually based on comparing the declared radionuclide inventory of the waste with the presumed or expected composition, which is estimated, based on the known history of the waste and its processing. The difficulty of such estimations for radioactive components from nuclear fuel assemblies is that reactor design parameters and operating histories can have a significant influence on the nuclide inventory of any individual fuel assembly. Thus, knowledge of these parameters is a key issue to determine the realistic concentration ranges, or bandwidths, of the radionuclide inventory. As soon as a governmental decision on the construction of a high-level waste repository will be made, comprehensive radionuclide inventories of the wastes assigned for the deposition will be required. The list of final repository relevant radionuclide is based on the safety assessment for this particular repository, thus it is likely to comprise more-or-less the same radionuclides that need to be declared for compacted metallic waste residual from the reprocessing of spent fuel assemblies. In Germany, the radionuclide declaration list for the disposal of used fuel assemblies is not yet specified. An estimation of the average radionuclide composition of the burnt-up fuel including the realistic inventory bandwidths for each of relevant radionuclides would be highly desirable beforehand. This information is needed for the development of proof tools for the product quality control or safeguards, but also for the evaluation of various safety scenarios regarding the radionuclide mobility or contamination. This work is focused on the development of a method for the determination of realistic radionuclide bandwidths in cases when no information of reactor design and operating data is available. Reactor parameters are classes as Primary Reactor Parameters of burn-up (BU) and cooling time (CT) that are considered to be known, and so-called Secondary Reactor Parameters (SRPs) that include nine parameters that are analysed: initial enrichment (IE), fuel density (FD), fuel temperature (FT), specific power (SP), downtime (DT), irradiation time (IT), moderator density (MD), moderator temperature (MT) and boric acid concentration (BA) used in the water for reactor control. The modelling of radionuclide inventories is carried out with the burn-up code SCALE 6.1 using the nuclear data library ENDF/B-VII.0. The input data include geometry of the fuel assembly and a set of the associated SRP values. The magnitude of the bandwidth significantly varies for different radionuclides and depends strongly on the primary parameters of burn-up and cooling time. The theoretical bandwidths are validated with experimental data. For this purpose the destructive radiochemical assay (RCA) data are taken from the Spent Fuel Isotopic Composition Database (SFCOMPO), which is maintained by the OECD Nuclear Energy Agency. There is, however, presently insufficient experimental data to validate the bandwidths for all radionuclides of interest to waste management. It is however possible to perform a comprehensive verification by comparing the bandwidths obtained with other reference inventory calculations used by several countries for long-term safety assessment, spent fuel transportation and storage, or other applications This work provides realistic radionuclide bandwidths that support methods for a long-term safety analysis of final repositories as well as for the development of efficient and validated tools for the PQC of HLW disposal and for safeguard applications.