Traditional engineering and physics frameworks rely heavily on continuum mechanics. When modeling how a solid steel beam bends or how water flows through a pipe, we assume the material is continuous and unbroken. We divide the medium into a smooth, connected mesh of elements and solve partial differential equations across them.
But this assumption breaks down completely when materials fracture, flow, or jam. If you try to model a rotary blade cutting through soil, a rock mass shattering during an explosion, or concrete spalling under acid rain, a continuous mesh cannot easily cope with the sudden appearance of infinite new surfaces and disconnected fragments. As Wusong XIAO et al. observe in SIMULATION ANALYSIS AND OPTIMIZATION OF SOIL CUTTING OF ROTARY BLADE BY ANSYS/LS-DYNA, traditional finite element methods struggle to simulate highly discontinuous behaviors like splashing and breaking.
To accurately capture these phenomena, we must abandon the continuum and treat the material as a collection of distinct, interacting pieces. This is the core philosophy of the Discrete Element Method (DEM). By tracking the individual trajectory and rotation of every single particle or block, as detailed by Masahiro Fukuda et al. in Numerical Analysis on Passage and Blockage Behaviors of Fine Particles through an Orifice Consists of Coarse Particles, DEM naturally captures macroscopic behaviors like clogging, sliding, and complete structural detachment from the ground up.
The Micro-Mechanical Foundation: Contacts and Forces
How does DEM simulate the motion of thousands or millions of independent particles without a unifying mesh? The answer lies in explicit, localized contact laws.
Instead of treating particles as perfectly rigid billiard balls that bounce instantly, DEM typically employs a "soft-sphere" approach. When two particles collide, they are allowed to overlap slightly. The simulation treats this tiny overlap as a proxy for physical elastic deformation. At each time step, the algorithm identifies all overlapping pairs and calculates explicit normal and tangential forces based on the magnitude and velocity of that overlap.
To build an accurate simulation, these contact laws must reflect real-world friction and energy loss. Eric DeGiuli in Friction law and hysteresis in granular materials utilizes standard DEM techniques incorporating Coulomb friction limits alongside linear elastic-dashpot interactions to demonstrate how mechanical noise emerges from stable contacts.
To capture more complex, non-linear deformations, researchers frequently implement the Hertz-Mindlin contact model. As explained by Cunliang Shen et al. in Effect of Packing Structure Evolution on the Flow Characteristics in a Binary Composite Packed Bed Based on DEM-CFD Method, this model accounts for both the non-linear elastic repulsion of spheres and the dissipative forces (damping) that govern energy loss during high-speed impacts. By combining these non-linear contact force laws with friction limit conditions, the simulation precisely tracks how forces are shared across large-scale particle systems (Chaoming Song et al. in From force distribution to average coordination number in frictional granular matter).
Cohesion and Fragmentation: Modeling Solids That Break
If DEM models materials as separate particles, how can it simulate a solid, cohesive material like a rock mass or a block of concrete? The solution is to introduce virtual "glue" between adjacent particles via bonded particle models.
By establishing cohesive bonds that resist tension, compression, and shear, a collection of discrete particles behaves exactly like a solid continuous medium under low stress. However, unlike traditional continuum models that require complex, ad-hoc re-meshing to show a crack, a bonded DEM model handles failure naturally. When the local stress between two particles exceeds a defined threshold, the bond simply snaps.
Guoxin Li in Propiedades mecánicas del hormigón corroído por la lluvia ácida con base en el elemento discreto de partículas (Editado por el Dr. Kimon Christianis, University of Patras, Greece) demonstrates that representing heterogeneous materials like concrete as an assembly of bonded particles allows the simulation to naturally capture material discontinuities, such as cracks and spalling, by calculating the independent motion and separation of individual aggregates.
For highly brittle materials where the interlocking shape of the grains dictates strength, researchers deploy advanced variants like the flat-jointed model. As Tingting Liu et al. explain in Effects of Stress States and Joint Configurations on Dynamic Mechanical Properties of Rock Masses, the flat-jointed model simulates the meso-mechanical behavior of interlocking particles and cohesive bonds, making it uniquely suited for capturing how rock masses deform and fracture under sudden dynamic impacts.
Once these bonds break, the material instantly transitions from a solid solid to a flowing granular mass. This makes DEM indispensable for capturing high-fidelity material fragmentation and transport dynamics that defy continuum mechanics, such as industrial cutting tools breaking through heterogeneous coal seams (Xin Jin et al. in Research on the optimization of drum structure based on virtual prototyping technology) or the propagation of cracks during deep-earth rock blasting (Ruizong Xu et al. in Research on the damage characteristics of rock masses based on double guide-hole blasting under high in-situ stress).
From Microscopic Contacts to Macroscopic Phenomena
The ultimate value of DEM is its ability to serve as a bridge between micro-scale contact physics and macro-scale material behavior. By aggregating the forces across thousands of individual contacts, researchers can observe how large-scale structural networks spontaneously emerge.
For instance, the macroscopic strength and friction angle of a granular mixture are deeply tied to its internal coordination number (the average number of contacts per particle) and how forces transmit through these contact networks. Kejia Wu et al. in Experimental and DEM Study on the Mechanical Behaviors of Sand–Fines Mixtures with Different Fines Contents and Particle Size Ratios demonstrate that the peak and critical strength of sand-silt mixtures depend on how strong and weak particle contacts cooperate within the network under stress.
Similarly, by systematically varying friction coefficients and preparation protocols, DEM simulations allow physicists to map out the phase space of jammed matter—exploring the exact boundary where a loose, fluid-like collection of grains suddenly locks into a rigid, solid disordered arrangement (Christopher Briscoe et al. in Entropy of Jammed Matter).
[Micro-Scale Input] [Meso-Scale Interaction] [Macro-Scale Phenomenon]
Overlap & Friction ---> Bond Snapping & ---> Flow, Jamming, Clogging,
Contact Laws Force Chain Networks & Total Fragmentation
This micro-to-macro mapping is so robust that its core principles are being extended far beyond geological and industrial materials. For example, Huijing Du et al. in Multiscale modeling of layer formation in epidermis represent biological cells as clusters of discrete elements governed by Lennard-Jones type potentials. They demonstrate that tracking discrete, localized forces can successfully simulate complex physical behaviors like cell growth, deformation, and division, showing how robust biological boundaries emerge from simple, local mechanical rules.
Key Foundational Work
From force distribution to average coordination number in frictional granular matter (2010) — contributed 1 concept elements to this direction.
Entropy of Jammed Matter (2008) — contributed 1 concept elements to this direction.