Study of the soil-interaction behaviour with underground structures under unsaturated conditions

Abstract

Soil-structure interface behaviour is an interesting topic due to the complexity of the interface mechanics. The interfaces between cohesionless soils and solid structure elements are encountered in different geotechnical engineering projects. The shear strength and stiffness characteristics, the thickness of the interfacial layer, the bonding and slipping properties are playing significant roles in an understanding the mechanical behaviour of such interfaces. One of the key parameters for the design and safety assessment of the engineering structures (e.g. retaining walls, deep and shallow foundations, tunnels and earth reinforcement) is the shear strength at the interface. Suction is an important stress-state variable of unsaturated soils. The magnitude of matric suction affects the shear strength and the volume change of soil and soil-concrete interfaces, thus the adequate characterization of interface behaviour is significant for its precise performance predictions. The prime aim behind this study is to investigate the behaviour of interface between compacted silty sand soil and concrete counterfaces at (a) different initial void ratios, (b) different surface roughness (smooth and rough) under the influence of different levels of applied vertical stress and test conditions (saturated and constant water content). The second main objective of this study is to investigate the effect of void ratio and the effect of the applied vertical stress level on the variation of matric suction during direct shear tests (matric suction stabilisation, consolidation and shearing stages). To do so, a new loading steel cap of large-scale direct shear apparatus for testing soilsoil and soil-concrete specimens has been manufactured. Firstly, a series of large-scale (300 mm x 300 mm) direct shear tests were carried out on compacted soil samples under different levels of applied vertical stress, void ratios and test conditions (saturated and constant water content). The experimental results confirm the dependency of shear strength, volumetric behaviour and measured matric suction on the vertical stress and initial void ratio. Secondly, to investigate the interface behaviour with different surface roughness (smooth and rough) and compare it with the behaviour of soil-soil samples, a number of interface direct shear tests were conducted between silty sand and a concrete pad under the same levels of vertical stress, void ratios and test conditions. The trend of behaviour of the shear strength versus horizontal displacement curves of soil-concrete interface tests is similar to those of soil tests. The laboratory tests results show that the surface roughness, vertical stress, void ratio and test conditions have significant influence on the shearing characteristics of the interface samples. The study noted that the strain softening behaviour of the tested material is noticeably influenced by the initial void ratio of specimens and surface roughness for both test conditions. From the results, it was observed that the initial matric suction has a clear dependency on the void ratio of the specimens. The shearing behaviour of the soil samples was higher than the rough and smooth interfaces for both studied void ratios and test conditions, whereas, the smooth interface showed lower values of shear strength of all the tested samples. It was noted that there is a remarkable decrease in matric suction during shearing stage with the level of applied vertical stress and the most important matric suction evolution was occurred before the horizontal displacement corresponding to the peak/maximum shear strength achieved.