A number of key neutronics analyses were performed for a range of reactor core states, including the beginning-of-life, middle-of-life, and end- of-life states. These studies included analyses of:

• Power distributions (for use in the thermal/hydraulic analyses), in­cluding (1) total fuel assembly power and core power distributions; and (2) axial and radial power distributions in the maximum power fuel assembly;

• Shutdown margins as a function of fuel burnup; and

• Key reactivity parameters, including (1) delayed neutron fraction[54]; (2) prompt neutron lifetime[55]; (3) control element worth[56]; and (4) prompt temperature coefficient.[57]

The neutronics of the reactor core were modeled using Los Alamos National Laboratory’s Monte Carlo n-Particle code, version 5 (MCNP5) with the core nuclear reaction database ENDF/B-VII maintained by the National Nuclear Data Center. In addition, Argonne National Laboratory’s REBUS codes for analysis of fuel cycles were used for the burnup analysis. Finally, some confirmatory analysis was performed using the HELIOS two­dimensional generalized-geometry lattice physics transport code.[58]

Several challenges were associated with performing these analyses at Wisconsin. First, sufficient information was not available on the operational history of the HEU core to be able to calculate fuel composition for use in benchmarking the model. As a substitute, analysts worked backwards to estimate the composition of the fuel using the current critical conditions for the core. This does not provide a benchmark but gives some confidence in the validity of the model. Second, large computing resources were required for some of the analyses, beyond what was easily available at the university. Finally, the university had only a modest existing capacity for performing such reactor analyses. This capacity had to be built up for the analyses to be carried through successfully.

The major difficulty associated with conversion was related to system reactivity. The like-for-like replacement of HEU-FLIP fuel with LEU 30/20 fuel increased the reactivity of the core. The modeled core with LEU fuel could not be shut down even with all control elements fully inserted. To reduce system reactivity and meet shutdown margin requirements, the core design was changed from a 23 fuel assembly/10 reflector configura­tion (in which the assemblies are arranged in an “H” pattern) to a 21 fuel assembly/14 reflector configuration (in which the assemblies are arranged in an “X” pattern) (see Figure 3-1). However, the reduction in the number of fuel assemblies resulted in a slightly reduced core lifetime following conversion.[59]