Effect of matric suction on pile behavior in unsaturated soils by physical modeling in FCV

Deep foundations are often used in a typical engineering project to transfer loads from heavy superstructures (bridges, highways, embankments and high-rise buildings) to the subsoil safely without subsidence or instability problems. In order to achieve an accurate and economical design in each of these projects, we must have a good understanding of how deep foundations behave, which can be done through analytical methods, numerical methods and experiments or experimental observations. Since a series of simplifications are usually performed in numerical and analytical methods, the use of experimental methods is selected to achieve a more desirable result. Experimental methods can be done by examining real models on site or performing experiments on a small-scale model in the laboratory (physical modeling), which due to lower cost, simplicity and reproducibility, usually physical modeling compared to Real on-site models are preferred. For physical modeling, various devices such as Simple Chamber (1 g), Calibration Chamber (CC) and centrifuge device can be used. The limitations and difficulties of these devices led to the recent introduction of a device called the Frustum Confining Vessel (FCV) for the physical modeling of deep foundations. In this device, which is in the form of an incomplete cone, by applying pressure to the bottom of the device, a linear stress gradient can be created along the central axis to be consistent with what is actually happening. Despite the many advantages that this device , it does not have a system for measuring the matric suction in unsaturated soils. Therefore, in this article, by embedding precision instruments in the sample made of this device in Amirkabir University of Technology (FCV-AUT) and implementing and loading of a pile in the soil inside it, It is possible to study the behavior of piles in unsaturated soil and the effect of moisture and matric suction on the load-displacement curve. The pile used in this research is an open-end pile which was implemented by driving method in Firoozkooh silty sand with gravimetric moisture contents of 0, 5, 10, 15 and 20% and in each case a load-displacement curve was drawn. The floor pressure applied in this research is equal to 100 kPa, Which was determined by placing several stress sensors at different soil depths and measuring the stress changes in depth. The following results were obtained after experiments: In the tested soil, up to 5% moisture content, no significant lateral friction is observed between the pile and the soil, and the end resistance, which is the result of plugging the soil inside the pile, provides most of the bearing capacity of the pile. From 5 to 15% moisture content, the lateral friction between the pile and the soil is gradually mobilized and while creating an initial jump in the curve of load-displacement, it increases the bearing capacity of the pile exponentially. But from 15% moisture content onwards, the initial slope decreases again due to the reduction of lateral friction. Considering the final bearing capacity of the pile as the load corresponding to 10% of the displacement of the pile head, the effect of matric suction on the changes of final bearing capacity was investigated. It was found that By increasing the amount of matric suction, first the final bearing capacity of the pile increases and after reaching a certain value, it decreases. The cause of this phenomenon is changes in wet soil behavior in different areas of the soil-water characteristic curve (SWCC) and is affected by various factors such as grain size, initial moisture content and soil density percentage.