American Nuclear Society
Home

Home / Publications / Journals / Nuclear Technology / Volume 204 / Number 2

Thermal Conductivity Estimation of Compacted Bentonite Buffer Materials for a High-Level Radioactive Waste Repository

Seok Yoon, Min-Jun Kim, Seung-Rae Lee, Geon-Young Kim

Nuclear Technology / Volume 204 / Number 2 / November 2018 / Pages 213-226

Technical Paper / dx.doi.org/10.1080/00295450.2018.1471909

Received:December 27, 2017
Accepted:April 27, 2018
Published:October 10, 2018

A deep geological repository has been considered as one of the most appropriate options for the disposal of high-level radioactive waste (HLW), and it will be constructed in a host rock area at a depth of 500 to 1000 m below the ground surface. The geological repository system is based on the concept of an engineered barrier system, and it consists of a disposal canister with packed spent fuel, buffer material, backfill material, and intact rock. The buffer plays an important role to assure the disposal safety of HLW since it can restrain the release of radionuclides and protect the canister from the inflow of groundwater. Since an increased heat quantity is released from the disposal canister into the surrounding buffer material, the thermal conductivity of the buffer material constitutes a key parameter needed to analyze the entire disposal safety. Therefore, this study presents a thermal conductivity prediction model for compacted bentonite buffer material from Kyungju, which is the only bentonite produced in Korea. The thermal conductivity of the compacted bentonite buffer from Kyungju was measured using a hot-wire method according to varying degrees of saturation, dry density, and temperature. The measurements showed that the thermal conductivity was concurrently influenced by the degree of saturation, dry density, and temperature variation. A regression model was proposed to predict the thermal conductivity of the compacted bentonite buffer from Kyungju using the degree of saturation and the dry density as the dependent variables. An additional regression model was also introduced that incorporated the temperature variation as an additional dependent variable, and the two models were directly compared with each other.