Page 56 - Thermal Hydraulics Aspects of Liquid Metal Cooled Nuclear Reactors

Thermal-hydraulic challenges in liquid-metal-cooled reactors 31

Development needs
Within the near future, RANS models will remain the best way to predict the average behav-
ior of outlet jets in liquid-metal-cooled reactors. Hence, the development of these models and
their validation in the wide range of situations relevant to liquid-metal reactor thermal
hydraulics should be pursued. In situations where knowledge of fluctuations is required,
simulations coupling an overall RANS calculation to a local LES domain may provide a
less-costly alternative to full-scale LES simulations. Validation may be both experimental
(liquid-metal or simulant fluid) or numerical (against DNS simulations). Apart from that,
also the applicability of hybrid RANS-LES models could be studied.
Gas entrainment
l
Challenge
Gas entrainment is especially considered as a possible source of core voiding in a
sodium-cooled fast reactor. Because of its higher density, it is conceived less important
in a lead fast reactor. In order to assess the risk of gas entrainment, various gas source
terms must be assessed, in particular the possibility of bubble detachment from vortices
at the free surface. Then, the possibility of transporting these bubbles to the core inlet
plenum must be analyzed (see above). Finally, the potential for gas accumulation in this
inlet plenum must be assessed.

State of the art
The ongoing and state-of-the-art activities concerning gas entrainment are therefore care-
fully being performed within France. Tenchine et al. (2014) describe the status to this respect
in France in 2014. There are indications that the efforts in France in the meantime progressed
sufficiently suggesting that development and validation of experimental and numerical
approaches can be considered as established, allowing application of the developed and val-
idated methods for various designs.
Development needs
Small-scale analysis of steady-state vortices with two-phase CFD has been validated suc-
cessfully against simulant fluid and liquid-metal experiments. However, vortex formation
in liquid-metal reactor plena tends to be intermittent. Hence, ongoing efforts aim to extend
this small-scale analysis to unsteady vortices. Furthermore, a large separation of scales exists
in the reactor case between the overall flow driving vortex formation ( 10m) and the scale
of the vortices themselves ( 1cm): hence, the possibility of fully resolving the vortices
over the whole free surface remains unfeasible today. Instead, efforts have been concentrated
on the defining criteria able to predict whether a given large-scale flow (computed by a
coarse-scale simulation) could lead to the formation of gas-entraining vortices: the develop-
ment and validation of such criteria is ongoing.

Severe accidents:
l Corium cooling
Challenge
For the safety assessment of all nuclear reactors, a core disruptive accident that leads to a
partial or whole core meltdown needs to be considered (Rakhi and Velusamy, 2017). To
this purpose, various design concepts (like core catcher concepts either inside or outside
the reactor vessel) and safety measures have been suggested and incorporated in reactor
designs. The challenge is to assess the cooling capability of such design concepts and
safety measures.