Uncertainty Analysis of Advanced Fuel Cycles to Control Plutonium Inventories

Date of Award


Degree Type


Degree Name

Master of Science


Nuclear Engineering

Major Professor

Laurence F. Miller

Committee Members

Lawrence W. Townsend, Martin L. Grossbeck


This paper assesses the uncertainty associated with the utilization and implementation of advanced fuel cycles to control plutonium inventories. The specific fuel cycles investigated are a partially closed cycle utilizing MOX reactors and completely closed one-tier fuel cycles utilizing fast reactors. Multiple methods for assessing these uncertainties were utilized. A scenario approach that varied the time and number of the implementation of the advanced reactors was used. It was found that the implementation of 3 FR/yr with a CR of 0.5 could reduce the amount of Pu by over 36% in reference to building 3 LWR/yr. In addition to reducing the inventory with respect to the reference LWR case, the growth rate can be reduced from an initial 22 tons Pu/ year growth to 5 tons Pu / year growth with the 2030 actual initial Pu inventory implementation cases. The MOX cases keep the Pu/ TWhe inventory slightly above 1 ton Pu/TWhe and the extremely low CR FR cases even lower than that value. Thus from this work the extremely low CR FR scenarios show the greatest ability to control the growing Pu inventory. In addition to the scenario approach a Monte Carlo uncertainty model was developed and analyzed. The uncertainty analysis showed the high burn up cases are comparable with the of the low CR FR cases in there ability to control the Pu inventory with the Pu inventories ranging from 2500 tons of Pu to 7500 tons of Pu. However, for the high burn up cases the majority of the Pu is Out-Of-Pile as opposed to the FR cases where a considerable amount of the Pu is In-Pile. From a proliferation stand point, the low CR FR case is better at the controlling the Pu inventory because the total inventories are relatively the same for the majority of the runs, and the FR cases keep most of the Pu In-Pile rather than the high burn up cases which keep most of it Out-Of-Pile. Lastly, a brief economic uncertainty model was developed. The economic results show that the once-through cycle is the cheapest with over 50% of the test cases coming in cheaper than all of the FR and MOX cases. The FR cases come out to be the next cheapest with the MOX cases being the most expensive.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."