An experimental investigation on self-leveling behavior of debris beds using gas-injection

Studies on the self-leveling behavior of debris beds are crucial in the assessment of core-disruptive accidents (CDAs) that could occur in sodium-cooled fast reactors (SFRs). To further clarify this behavior, a series of experiments has been performed in which nitrogen gas has been percolated uniformly through particle beds. Although in the past, several experiments have been conducted to investigate this behavior, most of these were under comparatively lower gas velocities, the findings of which might be not directly applicable to actual reactor accident conditions. Current experiments were conducted in a cylindrical tank with the dimensions of 310 mm in inner diameter and 1000 mm in height, in which nitrogen gas, water and different kinds of solid particles, simulate the fission gas, coolant and fuel debris, respectively. During experiments, to accomplish the bubble-based leveling as expected in actual reactor conditions, two experimental approaches, termed respectively as the gas pre-charge method and the pressure-adjustment method, have been attempted. Through elaborate comparisons and evaluations, it is found that compared to the gas pre-charge way the pressure-adjustment method can alleviate the liquid disturbance from the bottom inlet pipelines more effectively throughout the whole experimental process. Further, based on the experimental observations and quantitative data obtained, influence of various experimental parameters on the self-leveling, including gas flow rate, water depth, particle size, particle density as well as boiling mode is checked and compared. From the analyses, it is also observed that, the convection in the water pool, though is not so evident for previous studies at lower gas velocities, is newly identified to play a significant role in this work, especially for the experimental runs using larger-size lower-density particles at comparatively higher gas flow rates and water depths. Current work provides critical evidence and a large palette of favorable data for a better understanding of CDAs and improved verifications of computer models developed in advanced fast reactor safety analysis codes. (c) 2013 Elsevier Inc. All rights reserved.