With the advent of modern microstructural testing techniques and microstructure based constitutive models the microstructural characterisation of soils is gaining prominence. This paper reviews the history of microstructure investigation in unsaturated soils and discusses the engineering significance of this research to date. After a brief overview of the main microstructural techniques, the paper focuses on the evaluation of the current state of use and the development of two widely used techniques to study the microstructure of partially saturated soils, namely mercury intrusion porosimetry and the environmental scanning electron microscopy. The details of these techniques, their advantages and limitations, are first covered, followed by the presentation of selected test results. These results highlight the use of these techniques for understanding different hydro-mechanical behavioural features observed at macroscopic scale. Specifically, the paper shows the use of these techniques to explore the fundamental properties of water retention characteristics, water permeability, and micro and macrostructural interactions along different hydro-mechanical paths.

No abstract

The soil-water characteristic curve (SWCC) of fine-grained soils is usually determined experimentally. In many applications, such as design of mine waste covers and landfill liners, the unsaturated permeability function, k(h), is often derived theoretically from the measured SWCC. Implicit in these derivations is the transformation of the SWCC to a pore-size distribution (PSD), typically assumed to be constant and mono-modal. PSDs of a clayey till compacted at various water contents were measured after compaction, after flexible-wall permeability testing, and during and after SWCC tests. The measurements show that the PSD changes significantly during permeability and SWCC testing. A method is advanced for predicting the observed changes in PSD during SWCC testing. PSDs are determined for soil samples subjected to the highest and lowest suctions applied during the SWCC test. The measured PSDs are transformed to account for pore trapping; the transform assumes that flow occurs through two sets of randomly distributed pores in series. To model pore shrinkage, the pores are idealized as elastic cylinders. PSDs measured after different suction applications in the SWCC tests are compared with predictions of the shrinkage model. The method can also be used to predict the SWCC. Measured and predicted values are compared.Key words: landfill liners, mine waste covers, soil-water characteristic curve, pore-size distribution.

Accurate predictions of drying rates are desirable to optimize surface deposition of thickened or paste tailings. A series of laboratory and field trials were implemented to study evaporation from tailings at the Bulyanhulu gold mine and were compared with numerical simulations using the unsaturated flow model SoilCover. The laboratory tests included two “large-scale” experiments on 10 cm thick layers of tailings 2 m by 1 m in plan, and a smaller column test on a 20 cm thick and 20 cm diameter sample. Data monitored during these tests included albedo, volume change, degree of cracking, matric suction, water content, and drying rate. Field data included gravimetric water contents and albedo values. The model could reasonably simulate the laboratory experiments when adjustments were made to account for self-weight consolidation and the effect of volume change on the relative permeability function. The model could simulate drying in the field for up to 3 weeks after deposition before the accumulation of gypsum and magnesium sulphate salts began to affect evaporation. Cracking and salt accumulation were observed both in the laboratory and in the field. A general model for simulating drying from paste tailings should incorporate the effects of cracking and salts.

No abstract