Modelling Hydraulic and Swelling Pressure Characteristics of Bentonite Buffer Material for High-Level Nuclear Waste Containment Conditions

High-level nuclear waste management is crucial from the environmental safety and health perspectives. The high-level nuclear waste consists of radioactive elements in a slurry state under high temperatures, which is encapsulated in a metal canister and then buffered with heavily compacted bentonite bricks. Therefore, it is important to study soil hydraulic and swelling characteristics of unsaturated bentonite under harsh chemical and high mechanical loadings. In this study, a simple bipoint model was proposed for the determination of wetting soil water characteristics curve (SWCC) under isochoric conditions for various bentonites. The drying soil water characteristics curve of bentonites was obtained theoretically using volumetric shrinkage data. Apart from the soil–water characteristics, swelling pressure is a key factor in the design of a nuclear waste repository. Therefore, a machine learning model was proposed for the estimation of swelling pressure of bentonites using basic clay-electrolyte characteristics. The proposed models were useful for quickly assessing the drying–wetting SWCCs of various bentonites and Decision Tree machine learning algorithms was suitable for the estimation of swelling pressure with a mean absolute error of 0.78 MPa and coefficient of determination of 0.92.