Search for: All records

Temperature variations in low permeable soil (e.g. clay) induce pore pressure, which is known as thermal pressurization. Previous research showed that thermal pressurization highly depends on thermal pressurization coefficient. This coefficient depends on the soil type and changes with temperature due to temperature dependency of thermal expansion coefficient of water. Thermal pressurization is often investigated through thermo-hydro-mechanical (THM) numerical modeling. THM process, with respect to thermal loading, has been examined in the literature to justify the field observations by incorporating advanced thermo-mechanical constitutive models. However, result of numerical simulations using advanced thermo-elastoplastic models still show some discrepancies with experimental and field observations. In this study, the assessment of thermal pressurization in Boom clay is scrutinized through employing a relatively simple while practical thermo-poroelastic finite element model with careful consideration of the temperature-dependent thermal, hydraulic, and mechanical properties of the medium and saturating fluid (i.e. water). The numerical model is carried out using COMSOL Multiphysics and the results of the numerical simulations are compared and validated with the ATLAS project, a large-scale experimental facility in Belgium. The results confirm that thermal and hydraulic coupling parameters are the key factors to change thermal pressurization. 

The exact heat transfer mechanism in the soil media can be understood by analyzing the soil behavior surrounding the heat sources. In literature, heat conduction has been considered as a main heat transfer mechanism in soil, and less attention has been given to the natural heat convection in saturated soils. There is only limited research in the literature which shows the presence of thermally induced pore fluid flow in soil media. It has been observed that heat convection through pore fluid flow in sand facilitates heat transfer in the ground. Therefore, both heat conduction and heat convection must be considered to accurately model the heat transfer mechanism in soil. In this paper, the presence of natural convection of water in a 2D axisymmetric domain of soil with a vertical heat source has been numerically investigated in steady-state condition. The soil thermal response and heat transfer mechanism for different soil types are compared. Feasibility of thermally-induced pore fluid flow is analyzed for different soil types. The results determine the presence of thermally driven pore fluid flow in high permeability soil (e.g., coarse sand) and confirm that the effect of heat convection in low permeability silt and clay is negligible. 

The prediction of coupled nonisothermal multiphase flow in porous media has been the subject of many theoretical and experimental studies in the past half a century. In particular, the evaporation phenomenon from the shallow subsurface has been extensively studied based on the notion of equilibrium phase change between liquid water and water vapor (i.e., instantaneous phase change). One of the frequent assumptions in equilibrium phase change approach is that liquid water is hydraulically connected throughout the vadose zone. Furthermore, classical soil‐water retention curves (e.g., van Genuchten model), which have been extensively used in the literature to model evaporation process, are only valid for high and intermediate saturation degrees. Although these limitations have been addressed and improved in separate studies, they have not yet been rigorously incorporated in the numerical modeling of nonisothermal multiphase flow in shallow subsurface of in‐field soils. Therefore, the aim of this study is to investigate the coupled heat, liquid, and vapor flow in soil media through the Hertz‐Knudsen‐Schrage (HKS) phase change model and by incorporating a water retention model which captures the soil‐water characteristics from full to oven‐dried saturation degrees. A numerical model is developed and validated against the in‐field experimental data. Reasonable agreements between the calculated and measured values of water contents at all depths, as well as the temperature, and cumulative evaporation are observed. Results also confirm that the contribution of the film flow in overall mass flow in the medium is required for accurate modeling and cannot be ignored.