Evaluation of Flux- and Depletion-Based Methodologies for Tritium Breeding Ratio Prediction: Limitations of Current 7 Li Nuclear Data Processing

Tritium breeding ratio (TBR) is a key performance metric for fusion reactor blankets. This work compares flux-based and depletion-based methodologies for TBR prediction. The flux-based methodology calculates tritium production directly from neutron flux tallies and tritium-production cross sections, whereas the depletion-based methodology determines TBR from the tritium inventory generated during depletion calculations. This study evaluates both approaches using Monte Carlo neutronics models of the ARC fusion reactor implemented in MCNP 6.3, Serpent 2.2.1, and OpenMC. The flux-based methodology produced consistent TBR values across all three Monte Carlo codes. Comparison with depletion-based calculations revealed a systematic discrepancy associated with tritium production from 7Li. Further investigation showed that the aggregate tritium-production reaction for 7Li, 7Li(n,Xt), is obtained by summing inelastic cross sections rather than by representing a true 7Li(n,t) channel, which is not properly treated in current depletion calculations. As a result, tritium production from 7Li is not accounted for in the depletion calculations. This work further examined the feasibility of correcting this deficiency through a review of the underlying ENDF evaluation and identified additional complications associated with the breakup-flag and pseudolevel representations used in the original 7Li evaluation. Based on these findings, recommendations are provided for selecting the appropriate TBR methodology depending on blanket composition evolution and the significance of the 7Li contribution to tritium production. For systems with evolving compositions and significant 7Li contributions, a hybrid methodology combining depletion calculations with flux-based TBR evaluation is recommended.