Nuclear waste Educator's workshop: What and how do we teach about nuclear waste?

A multidisciplinary framework from reactors to repositories for evaluating spent nuclear fuel from advanced reactors

This study presents a multidisciplinary reactor-to-repository framework to compare different advanced reactors with respect to their spent nuclear fuel (SNF) disposal. The framework consists of (1) OpenMC for simulating neutronics, fuel depletion, and radioactive decays; (2) NWPY for computing the repository footprint given the thermal constraints; and (3) PFLOTRAN for simulating radionuclide transport in the geosphere to quantify the repository performance and environmental impact. We first perform the meta-analysis of past comparative analyses to identify the factors that led previously to their inconsistent conclusions. We then demonstrate the new framework by comparing five reactor types. Our analysis highlights the granularity and the specificities of each reactor and fuel type so that we should avoid making sweeping conclusions about advanced reactor SNF. Significant findings are that (1) the repository footprint is neither linearly related to SNF volume nor to decay heat, due to the repository's thermal constraint (2), fast reactors have significantly higher I-129 inventory, which is often the primary dose contributor, and (3) the repository performance primarily depends on the waste forms. The TRISO-based reactors, in particular, have significantly higher SNF volumes compared to the others but result in smaller repository footprints and lower peak dose rates. The open-source framework ensures proper multidisciplinary connections between reactor simulations and environmental assessments, as well as the transparency/traceability required for such comparative analyses. It aims to support reactor designers, repository developers, and policymakers in evaluating the impact of different reactor designs, with the ultimate goal of improving the sustainability of nuclear energy systems.

Machine learning surrogates for surface complexation model of uranium sorption to oxides

The safety assessments of the geological storage of spent nuclear fuel require understanding the underground radionuclide mobility in case of a leakage from multi-barrier canisters. Uranium, the most common radionuclide in non-reprocessed spent nuclear fuels, is immobile in reduced form (U(IV) and highly mobile in an oxidized state (U(VI)). The latter form is considered one of the most dangerous environmental threats in the safety assessments of spent nuclear fuel repositories. The sorption of uranium to mineral surfaces surrounding the repository limits their mobility. We quantify uranium sorption using surface complexation models (SCMs). Unfortunately, numerical SCM solvers often encounter convergence problems due to the complex nature of convoluted equations and correlations between model parameters. This study explored two machine learning surrogates for the 2-pK Triple Layer Model of uranium retention by oxide surfaces if released as U(IV) in the oxidizing conditions: random forest regressor and deep neural networks. Our surrogate models, particularly DNN, accurately reproduce SCM model predictions at a fraction of the computational cost without any convergence issues. The safety assessment of spent fuel repositories, specifically the migration of leaked radioactive waste, will benefit from having ultrafast AI/ML surrogates for the computationally expensive sorption models that can be easily incorporated into larger-scale contaminant migration models. One such model is presented here.