Shear behaviour of sand in Bi-directional direct simple shear tests

In geotechnical engineering, soil is subject to shear stress, as well as normal stress. In many cases, soil is sheared along multiple directions, such as an embankment under earthquake loading and foundation soil of a structure under a complex loading. In recent years, significant research has been devoted to understanding the static and cyclic shear behaviour of sand under complex stress conditions. To investigate the shear behaviours of soils, a few testing devices have been specifically developed. The most commonly used apparatuses are direct shear, direct simple shear, and hollow cylinder apparatuses. While each has its own merits, they share one common limitation. That is the soil specimens are sheared along only one direction in these devices, making it impossible to study the soil responses under multiple shear stresses encountered in many geotechnical engineering problems. It is widely acknowledged that testing stress path has a significant effect on shear stress, so specimens must be examined under a stress path similar to in-situ. In many cases, a simple shear apparatus closely duplicates the stress conditions in soil elements, and a bi-directional direct simple shear apparatus can create complex stress conditions. To investigate the shear behaviour of soils subject to complex loading conditions, several studies of multidirectional simple shear testing were performed on the first commercially available Variable Direction Dynamic Cyclic Simple Shear (VDDCSS) system. In the VDDCSS, the secondary shear actuator acting at 90 degrees to the primary shear actuator enables it to add shear stress in any horizontal direction. Various previously unexplored complex stress paths were tested in this study using the VDDCSS. Sand samples of Leighton Buzzard sand (Fraction B) were first subjected to consolidation shear stresses under drained conditions along different directions (from 0° to 180° to the X direction of the apparatus), followed by monotonic or cyclic secondary shear stresses along 0° until failure occurs. The magnitude and direction of the consolidation shear stress on the static and cyclic secondary shear behaviours of sand were systematically studied. In undrained static tests, soil strength was the lowest when the angles between the two shears stresses were near 90°, and the strength was the highest at 0°. In addition, a smaller angle produced a more brittle response, and a greater angle led to a more ductile response. The effect of stress path (the direction of consolidation shear stresses) was found to be more significant in tests with a greater magnitude of consolidation shear stresses, and the relationship between the angles (between the first shear stresses and secondary shear stresses) and shear behaviours was much more complex when the magnitude of consolidation shear stresses was increased. In drained static tests, the evidence of non-coaxiality was obtained. The non-coaxiality was the greatest at the initial stage of shearing, and it decreased to zero at higher shear strains. The degree of non-coaxiality was affected by the relative density of the specimen, vertical stress, level and direction of consolidation shear stress. In addition, the non-coaxiality was significantly affected by the consolidation shear stress, and the effect increased at a greater magnitude of consolidation shear stresses. The non-coaxiality increased as the angle was increased. In stress-controlled undrained dynamic tests, there was no significant difference in pore water generation rate among samples with consolidation shear stresses in different directions. However, samples with different directions of consolidation shear stresses failed at different numbers of cycles and in different directions. Liquefaction resistance was decreased by the increased magnitude of consolidation shear stress. In most of the tests with consolidation shear stress, the levels of liquefaction resistances were lower than that without consolidation shear stress. In strain-controlled undrained dynamic tests, liquefaction resistance (shear strength) was decreased from 0° to 90°, and increased from 90° to 180°. Liquefaction resistance in tests with consolidation shear stresses were lower than those without consolidation shear stress