ABSTRACT Interparticle cohesive forces play a crucial role in governing the shear behavior of many soils; however, quantifying this effect remains challenging due to limited microscopic insights available from laboratory experiments. In this context, the current study aims to investigate the influence of microscale cohesion on the shear response using the discrete element method (DEM), where the relevant stress, strain, and dilation characteristics are compared with experimental data before conducting a detailed micromechanical analysis. The results successfully replicate important features of a moderately compacted soil under shearing such as strain softening and contraction‐dilation responses. Interestingly, cohesion‐induced samples exhibit a sustained increase in shear stress even at high shear strain levels; for instance, the residual shear stress in the highly cohesive sample increased by approximately 40% compared to the non‐cohesive sample. This occurs because particle‐scale cohesion enhances interparticle bonding, leading to higher average contact numbers (coordination number CN) and increased skeletal forces. A key novelty of this study lies in establishing a quantitative correlation between conventional macroscale shear parameters and microscale CN under varying loading and cohesion conditions. This proves that the more cohesive the particles, the stronger the bonding effect with higher CN, consequently resulting in higher shear resistance. Moreover, this trend is exacerbated when the rolling friction of particles is introduced as the coupled effect of microscale cohesion and rolling friction strengthens the soil's resistance to shear deformation.
Building similarity graph...
Analyzing shared references across papers
Loading...
Doan et al. (Wed,) studied this question.
www.synapsesocial.com/papers/69d896046c1944d70ce073ce — DOI: https://doi.org/10.1002/nag.70319
Thao Doan
Buddhima Indraratna
Thanh Trung Nguyen
International Journal for Numerical and Analytical Methods in Geomechanics
University of Technology Sydney
University of Wollongong
Building similarity graph...
Analyzing shared references across papers
Loading...