화학공학소재연구정보센터
Powder Technology, Vol.336, 426-440, 2018
Modeling yield properties of compacted powder using a multi-particle finite element model with cohesive contacts
The multi- particle finite element method (MPFEM) was used to simulate the yield properties of compacted powders with cohesive contacts. A suitable representative volume element (RVE) of monodisperse, spherical, de formable particles was created to be implemented into a commercially available finite element code. New efficient periodic boundary conditions were proposed to compute representative properties of the volume at limited computational costs. A contact model was introduced, which includes repulsive forces, friction forces and cohesion forces. As a result, the proposed model is capable of considering tensile strength in a MPFEM setting, which was not attainable in related published work. We present extensive parameter studies to demonstrate the performance of the proposed RVE and to find the optimal balance between accuracy and computational speed. The minimum mesh fineness and the minimum number of particles in the RVE were determined during convergence studies. The employed explicit integration scheme was enhanced by means of mass scaling. The optimized model was used to predict the strength of compacted powders. A simple analytical expression was fitted to the numerical predictions to describe the uniaxial tensile strength and the uniaxial compression strength as a function of the powders' relative density and the cohesion strength of the contacts. A general form of a yield surface was proposed to describe the yield properties for generic load cases, which can be applied to different relative densities and cohesion strengths. As a result, we showed that the yield surfaces grow with increasing relative density, while they change their shape with increasing cohesion strength. The obtained yield surface results in the Drucker-Prager/Cap model in case of low cohesion, whereas it has an elliptical shape in case of high cohesion. The proposed analytical form of the yield surface is capable of describing both cases. (C) 2018 Elsevier B.V. All rights reserved.