Gap-graded soil, characterized by the absence of certain particle sizes, is commonly used in infrastructure projects such as dams and roadbeds. A comprehensive understanding of both the macro- and micro-mechanical behaviors of discontinuously graded soils is essential for their effective use in engineering applications. In this study, drainage triaxial compression tests were conducted on four gap-graded soil samples with different fine-grain contents mainly using the DEM method, whereas the flexible boundary part was performed using the FDM-DEM method. The contacts were classified based on the magnitude of contact forces between coarse and fine particles, considering the coordination number of the particles involved and the normal angular distribution of these contacts. This classification enabled a detailed analysis of how fine particles contribute to stress transmission and structural evolution during shearing. The fabric tensor for these contact types provided further insights into the anisotropy of samples during shearing. On the microscopic scale, the evolution of contact numbers was found to closely align with the observed stress-strain behaviors. Increasing fine particle content significantly altered the role of fine particles in the stress transmission process. With low content of finer particles, initially, fine particles were situated within the voids formed by coarse particles, and the fine particles are gradually embedded into the coarse particles during the loading process. With the increase of fine particle content, fine particles constantly aggregate to block coarse particles and become the main medium of stress transmission.

Discrete element simulation of triaxial tests is an important tool for exploring the deformation and failure mechanisms of geotechnical materials such as sands. A crucial aspect of this simulation is the accurate representation of lateral boundaries. Using coupled finite difference method (FDM)-discrete element method (DEM) approach, numerical simulations of consolidated-drained and consolidated-undrained triaxial tests were conducted under flexible lateral boundary conditions. These results were then compared with those of corresponding triaxial tests using rigid lateral boundaries. The results indicate that, compared to the rigid lateral boundary, the triaxial test using the FDM-DEM coupled flexible lateral boundary better captures both the macroscopic mechanical response and the microscopic particle kinematics of laboratory triaxial specimens. In the consolidated-drained triaxial tests, the strain softening and shear dilatancy of the specimen with the flexible lateral boundary are significantly weaker after reaching peak strength than those of the specimen with the rigid lateral boundary. In the consolidated-undrained triaxial tests, when the axial strain is large, the specimen with the flexible lateral boundary exhibits both a lower deviator stress and a smaller absolute value of negative excess pore pressure. Furthermore, in the consolidated-undrained triaxial tests, as the axial strain increases, the flexible lateral boundary provides weaker lateral constraint and support to the specimen compared to the rigid lateral boundary. Consequently, the stability of the force chains in the specimen with the flexible lateral boundary is lower, leading to more buckling events of force chains within the shear band. As a result, both the anisotropy and the deviator stress are reduced.

Triaxial tests are performed for remolded, artificially isotropic, and anisotropic structured samples under undrained conditions at confining pressures of 25, 100, and 200 kPa. Based on these test results, a binary-medium constitutive model is formulated based on homogenization theory and a breakage mechanism to describe the behaviors of structured soils. In this model, the binary-medium material is idealized as a representative volume element (RVE) composed of bonded elements, whose mechanical behaviors are expressed by the linearly elastic model, and frictional elements, whose mechanical behaviors are described by the double-yield surfaces constitutive model. The parameters of the bonded and frictional elements are determined from the test results of structured and remolded samples, respectively. The expressions for the breakage ratio and local stress coefficient matrix are introduced, and their parameters are provided. The computed results are compared with the test results, demonstrating that the model can reflect the main deformation features of structured soil relatively well, including the influence of anisotropy, gradual damage to particle bonding, and pore development.