Estimation of Effective Thermal Conductivity of Spherical and Ellipsoidal Shaped Randomly Packed Mono-Sized, Binary-Sized, and Poly-Dispersed Ceramic Pebble Beds

Lithium ceramics in the form of pebbles are chosen as one of the tritium breeding materials for the blanket of fusion reactor. The packing structure of the pebble bed (packed pebbles with gas inside the voids) influences its thermal properties, which are essential for the design of the breeder blanket units. In this study, simulations are conducted to obtain the effective thermal conductivity of the pebble beds. Discrete element method (DEM) is used to generate pebble beds for the study. The pebbles are generated in the funnel situated at the top of the container. They are subjected to free fall under the influence of gravity, resulting in a randomly packed pebble bed. DEM simulations are carried out with and without vibration to investigate the effect of vibration on the packing of the bed. In an ideal scenario, the pebbles are supposed to be perfectly spherical, but in practical cases, the lithium ceramic pebbles are nonspherical in shape. The effect of nonsphericity on the effective thermal conductivity of the pebble bed is studied in this work. The nonspherical pebble is generated using a multisphere approach; three spheres are merged together to depict the ellipsoidal shape. This study estimates effective thermal conductivities of mono-sized (MS), binary-sized (BS), and poly-dispersed (PD) lithium metatitanate pebbles in a helium gas environment. In the case of spherical pebbles, two different approaches, i.e., shrinking and enlarging the pebbles, have been used to avoid point contacts for meshing. Experimental measurement of the effective thermal conductivity of alumina and Li2TiO3 pebbles was carried out in a helium gas environment using the transient hot wire technique to benchmark the simulation results.