Thermal conductivity prediction model for compacted bentonites considering temperature variations

• Published in: Nuclear engineering and technology 2021, 53(10), , pp.3359-3366
• Main Authors: Yoon, Seok, Kim, Min-Jun, Park, Seunghun, Kim, Geon-Young
• Format: Article
• Language: English
• Subjects: Compacted bentonite; Metamodel; Multiple regression analysis; Temperature variation; Thermal conductivity; 원자력공학
• ISSN: 1738-5733; 2234-358X
• DOI: 10.1016/j.net.2021.05.001

An engineered barrier system (EBS) for the deep geological disposal of high-level radioactive waste (HLW) is composed of a disposal canister, buffer material, gap-filling material, and backfill material. As the buffer fills the empty space between the disposal canisters and the near-field rock mass, heat energy from the canisters is released to the surrounding buffer material. It is vital that this heat energy is rapidly dissipated to the near-field rock mass, and thus the thermal conductivity of the buffer is a key parameter to consider when evaluating the safety of the overall disposal system. Therefore, to take into consideration the sizeable amount of heat being released from such canisters, this study investigated the thermal conductivity of Korean compacted bentonites and its variation within a temperature range of 25 °C to 80–90 °C. As a result, thermal conductivity increased by 5–20% as the temperature increased. Furthermore, temperature had a greater effect under higher degrees of saturation and a lower impact under higher dry densities. This study also conducted a regression analysis with 147 sets of data to estimate the thermal conductivity of the compacted bentonite considering the initial dry density, water content, and variations in temperature. Furthermore, the Kriging method was adopted to establish an uncertainty metamodel of thermal conductivity to verify the regression model. The R2 value of the regression model was 0.925, and the regression model and metamodel showed similar results.