Feasibility Evaluation on European Capabilities for ²³⁸Pu based Radioisotope Power Systems

Production of Pu-238 proceeds via neutron irradiation of Np-237, which is created as a by-product in nuclear fission reactors, with a typical production slightly less than 1000 g/tU. When reprocessing spent nuclear fuel as e.g. done in the ORANO plant at La Hague (France) via the plutonium uranium reduction extraction (PUREX) process, neptunium is partially co extracted with uranium from dissolved irradiated nuclear fuel but as of today, it is not further refined, but instead added to the high-level waste and vitrified. The PUREX process can in principle be modified for neptunium recovery, and reprocessing of civil spent fuel can thus provide an abundant source of Np-237. Neutron irradiation of separated Np-237 to produce Pu-238 is conceptually simple, but producing sizeable quantities of Pu-238 with acceptable isotopic purity, separating it from the host matrix in which it is generated, its further processing and encapsulation, poses formidable technological challenges. All plutonium isotopes of technological interest are extremely radiotoxic alpha emitters. The elevated specific activity of isotope (238) adds highly concentrated radiolysis issues in liquid phase processing, static charging problems during powder handling, and heat generation when forming solid samples to the normal challenges of handling highly radiotoxic materials. Also, the precursor material Np-237 is a radiotoxic alpha emitter with a lesser specific activity compared to Pu-238. Especially the very rich and weakly explored chemistry of Np is a challenge here. In the present contribution, options for the irradiation of Np-237 containing transmutation targets in the BR-2 high flux reactor of SCK CEN (Mol, Belgium) are presented and boundary conditions for the production of such targets are discussed. The principal technology steps are: Fabrication of Np-237 targets for neutron irradiation Production of Pu-238 by neutron irradiation of Np-237 targets Processing of irradiated targets and Pu/Np separation Conversion of separated Pu-238 to solid PuO2 pellets Several technology options exist in each of the steps listed above, and these are reviewed with specific attention paid to those options which have been brought to actual production stage in the past, or which are intensively pursued today. Two principal options stand out: mixed ceramic-metal (CERMET) based routes developed and implemented at the Savannah River Site between the early 1950's and late 1980's and ceramic based routes pursued a.o. by Oak Ridge National Laboratory that have recently gained more attention. Technology options chosen in the early days are of course not necessarily the ones which would today be preferred. The European nuclear technology also developed differently over the past decades than the US nuclear technology, particularly with respect to civil spent fuel reprocessing, Pu separation and (U,Pu)O-2 manufacturing for Light Water Reactor (LWR) application. The two front-end processes (CERMET and full-oxide) for Np target production have similarities with established industry-scale fuel manufacturing processes in Europe: CERMET processes are applied in Materials Test Reactor (MTR) fuel fabrication and full-oxide processes are the reference technology for power reactors. The full-oxide process is furthermore also implemented at industrial scale for mixed uranium-plutonium (MOX) oxide fuel manufacturing, which shares similar radiotoxicity concerns as for Np-targets. CERMET front-end processes are in Europe applied for uranium-based fuels. An assessment of Pu-238 production capabilities in the BR-2 reactor has shown that suitable core positions can be selected with sufficiently low by-production of unwanted Pu-236. Further assessment will be needed to evaluate which is the flexibility regarding core positions that guarantee limited production of Pu-236. Production rates of Pu-238 were evaluated for unperturbed flux conditions to deduce an upper boundary of possible production rates. Preliminary calculations under perturbed flux conditions were performed for a design option that yet has to be optimized to deduce a lower boundary. The unperturbed flux results showed a theoretical upper boundary for the transformation yield slightly below 20%, achieved after three cycles of 28 days each and 28 days downtime between each cycle. Two production campaigns (i. e. six cycles) can reasonably be foreseen per year. Prolonged irradiation reduces the Pu vector below acceptable quality. The perturbed flux calculations for an un- optimized target showed a lower boundary slightly above 5% under the same irradiation conditions. Actual production yields are expected to be closer to the lower boundary than the upper boundary. Assuming a loading of 3 kg of Np, one may thus expect 150 g Pu-238 for a single production campaign, or 300 g Pu-238 per year. Design optimizations are expected to improve this yield. Regarding the processing of irradiated targets, the principal concerns are waste-related. The dissolution stage for the NpO2 Al CERMET targets is particularly problematic and from this perspective full-ceramic NpO2 targets would be preferred. The anion exchange purification stage as historically applied at SRS could be replaced by solvent extraction with TBP, most probably in combination with additional purification of the products via ion exchange. It is believed that the TBP process will create less waste than an anion exchange process and therefore seems to be the preferred method. Experience with these processes is available in Europe. The decay of Pu-238 dominates the radiology of the plutonium generated from Np-237 irradiation. Decay parameters of a typical Pu-vector issued by Np-237 irradiation have been compared with those of LWR MOX for which much broader experience exists in Europe. Even compared to MOX, issued from high burnup UO2 and including the ingrowth of Am-241 equivalent to a time lapse of 2 1/2 years, the radiological parameters per unit mass of Pu issued from Np-237 irradiation are higher by about two orders of magnitude. This difference will have to be taken into account for the Pu conversion process and the PuO2 production process.