DEM simulation of bonded granular material. Part I: Contact model and application to cemented sand. (May 2016)

• Record Type: Journal Article
• Authors: Shen, Zhifu; Jiang, Mingjing; Thornton, Colin
• Is Part Of: Computers and geotechnics. Volume 75(2016)
• Page Start: 192
• Page End: 209
• Publication Date: 2016-05
• Subjects: Discrete Element Method -- Cemented sand -- Bonded contact model -- Bond geometry -- Bond failure criterion; Engineering geology -- Data processing -- Periodicals; Soil mechanics -- Data processing -- Periodicals; Rock mechanics -- Data processing -- Periodicals; 624.1510285
• DOI: 10.1016/j.compgeo.2016.02.007
• Languages: English
• ISSNs: 0266-352X

Abstract

Abstract: A three-dimensional bonded contact model is developed in this study to simulate the mechanical behavior of bonded granular material using the Discrete Element Method. Two types of bonded contact are identified, i.e. parallel bond contact where the bond material deposits at a physical contact (two particles being in contact) and serial bond contact where the bond material deposits at a virtual contact (two particles being close but not in contact). The strength and stiffness of the bond material are incorporated into a previous unbonded contact model where the complete interactions in the normal, tangential, rolling and torsional directions are included. A grain is simplified as a sphere and the bond material is idealized in shape as a cylinder with concave ends complementary with the surfaces of the two bonded spheres. The effect of such bond geometry is considered in terms of bond strength and stiffness. The bond failure criterion can capture the coupled effect of normal force, shear force, rolling moment and torque on bond breakage. Parametric studies in DEM simulations show that the bond material properties, the bond content and the bond distribution all influence the behavior of cemented sand. With an increase in bond content or decrease in confining pressure, the mechanical behavior of cemented sand evolves from dilation-dominant where the peak state coincides with the maximum dilation rate, to bond-dominant where the peak state coincides with the maximum bond breakage rate. Microscopic statistics shows that the contributions of shear force and rolling moment to bond failure are statistically equal, both significantly greater than the contribution of torque.