Capabilities for property assessment on different levels of the micro-structure of DEM-simulated cementitious materials

• Analogue DEM-based simulation methodology for cementitious materials is presented.
• Non-spherical particles can be considered on different scales.
• Concrete’s pore structure is properly assessed by an advanced exploration method.
• Fibrous nature of C–S–H is modelled by nano-globules conforming to observations.
• Estimated permeability versus saturation degree agrees with experimental data.

Realistic three-scale simulation by DEM of particulate materials like concrete is nowadays possible by fast developments in computer technology. This would be an attractive alternative for systematic physical experimenting that is more time-consuming, labour-intensive and thus expensive. The paper discusses the simulation on meso-level of the aggregate structure and on micro-level of the fresh binder material that is enclosed by aggregate grains. In the latter case, it will be followed by hydration simulation to obtain the matured material. This can be expanded to involve particle packing of the C–S–H on nano-level, completing the three-scale approach. For assessment of mechanical properties, packing density is a crucial issue that can be straightforwardly studied. A subject of major engineering relevance is porosity because of durability issues. This requires techniques for delineating the capillary pore network structure to be able assessing its topological and geometric properties. By combining such features with hydraulic properties in a so called tube network model, basically transport properties of concrete can be estimated that are underlying long-term strength issues. Influences of technological parameters on packing characteristics are of interest for optimising packing, strength and durability of cementitious composites. The paper will offer the main lines of a complete methodology involving a modern three-scale analogue DEM approach, hydration simulation, and pore delineation and measuring. Intension is not to discuss separate “building blocks” of the methodology, because they have been presented, discussed and validated earlier in publications. Additionally, typical capabilities of the methodology will be indicated. To do so, results will be presented for illustrative purposes on the shape issue involved in aggregate packing, on the positive effects of cement blending by fine-grained rice husk ash on pore characteristics and resulting permeability, and on permeability estimation in the practical case of pores containing capillary water.