Abstract
The tritium release behavior from LiAlO2 samples in the TEQUILA experiment was analyzed using the MISTRAL code. This was done in order to benchmark the code for analyzing the performance of a LiAlO2 blanket test section under ITER-like conditions. Material property data available from the experimental sample microstructure characterization and from the literature were used as input to the code. The microstructure characterization was quite thorough and included the pore size distribution which was used to estimate the pore diffusion coefficient. In the case of the bulk diffusion coefficient, since single crystal experimental measurements are not available, two different values from different experimental data were used.
The strategy was to model four different transients for the same sample and to use the property data, in particular the diffusion coefficient, which will better reproduce all four transients. The transients studied were: Two temperature transients, in which the temperature changed by +50-degrees-C and -50-degrees-C and two hydrogen concentration transients in the purge, in which the concentration changes from 0.1% to 1% and from 1% to 0.1%. The results showed that the assumed bulk diffusion coefficient can change the output substantially. For each case, the effects of other parameters, such as the adsorption activation energy and pore diffusion coefficient, were also considered. The results are discussed in the paper.