Discrete element modeling of sand behavior in a biaxial shear test

• Published in: Journal of Zhejiang University. A. Science Vol. 9; no. 9; pp. 1176 - 1183
• Main Authors: Huang, Zhi-yi, Yang, Zhong-xuan, Wang, Zhen-yu
• Format: Journal Article
• Language: English
• Published: Hangzhou Zhejiang University Press 01.09.2008; Department of Civil Engineering,The University of Hong Kong,Hong Kong,China; MOE Key Laboratory of Soft Soils and Geoenvironmental Engineering,Zhejiang University,Hangzhou 310027,China; Department of Civil Engineering,Zhejiang University,Hangzbou 310027,China%Department of Civil Engineering,Zhejiang University,Hangzbou 310027,China
• Subjects: Civil Engineering; Classical and Continuum Physics; Engineering; Industrial Chemistry/Chemical Engineering; Mechanical Engineering; 临界状态; 土力学; 土壤性质; 显微结构; Critical state; Discrete element method (DEM); Microstructure; Granular soil behavior; TU43; Discrete element method(DEM)
• ISSN: 1673-565X; 1862-1775
• DOI: 10.1631/jzus.A0720059

The mechanical behavior of sand is very complex, and depends on factors including confining pressure, density, and drainage condition. A soil mass can be contractive or dilative when subjected to shear loading, and eventually reaches an ultimate state, referred to as the critical state in soil mechanics. Conventional approach to explore the mechanical behavior of sand mainly relies on the experimental tests in laboratory. This paper gives an alternative view to this subject using discrete element method （DEM）, which has attracted much attention in recent years. The implementation of the DEM is carried out by a series of numerical tests on granular assemblies with varying initial densities and confining pressures, under different test configurations. The results demonstrate that such numerical simulations can produce correct responses of the sand behavior in general, including the critical state response, as compared to experimental observations. In addition, the DEM can further provide details of the microstructure evolutions during shearing processes, and the resulting induced anisotropy can be fully captured and quantified in the particle scale.