Nuclear waste reprocessing and nonproliferation models are needed to support the renaissance in nuclear energy. This report summarizes an LDRD project to develop predictive capabilities to aid the next-generation nuclear fuel reprocessing, in SIERRA Mechanics, Sandia's high performance computing multiphysics code suite and Cantera, an open source software product for thermodynamics and kinetic modeling.Much of the focus of the project has been to develop a moving conformal decomposition finite element method (CDFEM) method applicable to mass transport at the water/oil droplet interface that occurs in the turbulent emulsion of droplets within the contactor. Contactor-scale models were developed using SIERRA Mechanics turbulence modeling capability. Unit operations occur at the column-scale where many contactors are connected in series. Population balance models 4 were developed to investigate placements and coupling of contactors at this scale. Thermodynamics models of the separation were developed in Cantera to allow for the prediction of distribution coefficients for various concentrations of surfactant and acid. Droplet-scale modeling was conducted in a microfluidic device and for verification of the algorithm. Extensive validation and discovery experiments were performed at the droplet and contactor-scales for both fluid dynamics and mass transport.
5
AcknowledgmentsThe authors would like to thank our mission specialist Veena Tikare for helping us develop the project, our reprocessing material balance expert Ben Cipiti for explaining the details of the reprocessing plant., Roger Pawlowski and Rich Schiek helped with the idea of a scalable network model, but were unable to continue with the project, though were very helpful in the first year of the work. We believe this type of work should be continued at Sandia. Paul Galambos and John Pflug supported our microfluidic experiments. Reviewers Randy Schunk and Lisa Mondy provided helpful feedback. Randy Schunk also helped with our droplet-scale modeling in a moving reference frame. The Aria Product Owners past and present, Amalia Black, Sheldon Tieszen, and Kim Mish, have been invaluable for keeping the project going despite the many other directions of Sierra Mechanics. We would like to thank the "Enabling Predictive Simulations" investment area of the LDRD program for funding this work. We would like to express our gratitude to the LDRD office for the extension given for receiving the final report due to the illness of the PI.6