Speaker
Description
The Water-Cooled Lithium-Lead (WCLL) is one of the breeding blanket concepts proposed by Europe in view of its DEMO reactor. The multi-physics of the system, multi-material domains, and complex blanket geometry characterize some issues in the development of a tritium transport model. However, the prediction of tritium concentrations and inventories in the blanket and the quantification of the permeation rate from the lead-lithium into the coolant is of main interest, both to guarantee fuel self-sufficiency and from a safety viewpoint.
In the years, the research activity conducted by the group carried out to development of a component level analysis focused on a simplified geometry of WCLL (Fig. 1) relative to the reference breeder unit of the central outboard equatorial module. The analysis involved step-by-step the evaluation of temperature and velocity fields, including buoyance forces and magnetohydrodynamic (MHD) effect, with focus on tritium transport (advection-diffusion of tritium into the lead-lithium eutectic alloy, transfer of tritium from the liquid interface towards the steel, diffusion of tritium inside the steel, transfer of tritium from the steel towards the coolant).
The tritium concentrations and the inventories inside the lead-lithium, in the Eurofer pipes and in the baffle have been evaluated.
No experimental data are currently available to validate the model in its complexity. In this framework, a code-to-code comparison has been carried out in order to verify the model. The comparison has been carried out between tritium transport models developed by PoliTo and CEA groups. The comparison has been carried out in a representative 2D slice of the WCLL and accounts for computational fluid dynamics, heat transfer, diffusion effects and trapping effects. In table 1, the result on the spatial integration on the 2D slice, showing the Tritium inventory per meter of thickness for different materials and in different conditions. The comparison showed a difference of 17.1% in the inventory and 24.7% as the highest error on the trapped content in Eurofer. Most of the Tritium content is within the PbLi. In conclusion, coherence between the two modeling strategies and their results has been found for the employed input data.
Starting from these results, the 3D model has been enriched by including trapping phenomena in the WCLL geometry with MHD and buoyancy effects, as well as the first five pulses of a start-up transient has been examined too. The tritium balance in a WCLL breeding unit is shown in Fig.2. Results show two transients: a faster transient which shows the tritium accumulation in PbLi and a slower transient which shows the tritium accumulation in Eurofer and tungsten. It has been found that more than 90% of the tritium losses are imputable to the breeding zone cooling, but the first wall flux is increasing in time. Also, the fraction of lost tritium flux vs processed tritium in the outer fuel cycle has been evaluated to be ranging from 0.08% to 0.17%, but it is increasing with time following the slower transient which involves the accumulation of tritium in the structural material.
| Speaker's title | Ms |
|---|---|
| Speaker's email address | raffaella.testoni@polito.it |
| Country/Int. organization | Italy |
| Affiliation/Organization | Politecnico di Torino |
