Granular materials cover a broad area of research at the intersection of different scientific fields including soft matter physics, soil mechanics, powder technology, agronomic transformations, and geological processes. Despite the wide variety of physico-chemical and morphological grain properties, the discrete granular structure of these materials leads to a rich generic phenomenology, which is at the focus of granular physics.
During the last thirty years, the particle-scale numerical modeling of granular materials has become a powerful and reliable research tool of granular physics. This discrete approach is based on the integration of the equations of motion simultaneously for all particles, described as rigid elements, by considering the contact forces as well as external forces acting on the particles. Given the boundary conditions, the mechanical response of an assembly of particles to external loading leads to relative particle motions constrained by steric exclusions in a dense state and/or by inelastic collisions in a loose or dilute state.
The discrete approach requires a time-discretized form of the equations of motion governing particle displacements, on one hand, and a force law or a force-displacement relation, describing particle interactions, on the other hand. When P. Cundall applied this method to granular geomaterials, he called it distinct element method (DEM) in contrast with finite element method used in continuum mechanics. The attribute “distinct” refers to the degrees of freedom carried by individual particles, but it was later replaced by “discrete” as a way to underline the discrete nature of the system. A similar method called molecular dynamics (MD) was used at that time for the simulation of molecular systems with classical schemes that could be directly applied to granular media. For this reason, many authors keep using indifferently the acronyms MD and DEM for the discrete simulation methods of granular materials.
A new method, called contact dynamics (CD), was later introduced by Moreau and Jean. This method is based on nonsmooth dynamics, a mathematical framework that accounts in its formalism for unilateral constraints at the contact points and temporal discontinuities due to collisions and Coulomb friction. Hence, one must distinguish between MD-DEM and CD-DEM. In the same way, in the limit of dense systems, the dynamics may sometimes be replaced by a quasistatic (QS) treatment in which the evolution of the system is described by consecutive states of mechanical equilibrium. This is an interesting numerical approach, which can be applied to a broad class of problems and to which we refer as QS-DEM. Finally, dilute systems (granular gases) are governed by an event-driven (ED) dynamics in which the state evolves by consecutive binary collisions between particles. This method, to which one may refer as ED-DEM, has been technically refined for fast simulations of collisional dilute systems. All these DEM models and techniques share the basic feature of reproducing emergent marco-scale properties from particle-scale input.
All discrete-element methods were first developed for a minimalistic physical model of granular materials where the particles are either spheres or disks of nearly the same size (monodisperse) interacting via frictional contacts. However, real granular media involve a variety of particle shapes and size distributions, as well as rheological behaviors (brittle, plastic, cohesive) at their contact zones. The particles and contacts may also evolve while interacting with their chemical, physical and thermal environment (coupling) or undergo fragmentation and wear. A realistic modeling of such features within a DEM framework is a major topic of ongoing developments with potential applicability to a wide variety of natural and industrial processes. Still more advanced features of DEM concern particles embedded in a fluid or in a solid phase. To deal with such problems, discrete modeling should be coupled with appropriate allied methods or models for the interstitial phase.
DEM is the proper method for the investigation of granular systems. Even in its minimalistic formulation, a DEM simulation leads to emergent complex behaviors arising from nonlinear interactions among a large number of degrees of freedom. The detailed information provided by such simulations may be used to analyze the particle-scale origins of many granular phenomena, be in static equilibrium or in dynamics. The role of DEM is all the more crucial that there is presently no general theoretical tool for the prediction of all states and transformations of granular matter. Indeed, in the absence of thermal randomness, a granular system can be found in various states of packing and flow. Most research work has focused on particular reference states such as steady flows of dense random and isotropic states. But the physics of transitions between different states of density and texture, times effects and the influence of particle properties are still poorly understood.
This site is devoted to the computational physics of granular media. The authors of the site wish to bring together pedagogical presentations to various aspects of discrete numerical modeling with technical details often omitted from regular research papers. A numerical methodology requires not only a simulation method but also a “toolbox” of different methods for the construction of granular samples, the management of boundary conditions and the choice of parameters. We are also concerned with advanced developments dealing with complex particle shapes, cohesion forces, hydrodynamic and thermal interactions and modeling of complex granular systems. The site presents also examples of codes and algorithms as well as applications mainly for a better understanding of complex granular behaviors with the corresponding references for further reading. Starting with a first circle of authors, our ambition is to enrich this site until a critical mass of information is condensed before opening in a second step to the rest of the community.