TIARA ANALYSIS OF TRITIUM INVENTORY IN LI2O

The TIARA code has been developed to predict tritium inventory in Li2O breeder ceramic and to predict purge exit flow rate and composition u nder steady-state operating conditions. Inventory predictions are base d on models for bulk diffusion, surface desorption, solubility and pre cipitation. Parameters for these models are determined from the result s of laboratory annealing studies on unirradiated and irradiated Li2O and from a limited number (2) of inventory data measured after in-reac tor purge flow testing. The remaining inventory data points (18) are u sed for code validation. In the validation exercise, models and model parameters are fixed prior to running the validation cases to assess h ow well TIARA predictions agree with data. On average, the TIARA predi ctions are in excellent agreement with the inventory data from the fol lowing in-reactor tests: EXOTIC-2, SIBELIUS, VOM-15H, CRITIC-1, BEATRI X-II (Phase 1) thin ring, and BEATRIX-II (Phase 1) thick pellet. Thus, TIARA can be used with a reasonable degree of confidence for design a nalysis over a broad range of fabrication variables and steady-state o perating conditions.