LAUSR

Abstract Using meso-structural representations of concrete, with aggregates dispersed in mortar, has become a standard approach for damage and fracture analysis. However, there is no full agreement on appropriate modelling of different phases and on the inclusion of interfacial transition zones (ITZ). This work explores different mortar and ITZ formulations and by comparison with own experiments demonstrates that the optimal strategy balancing physical realism and computational efficiency requires: (1) damage-plasticity formulation for mortar, calibrated with mortar tension and compression experiments; (2) cohesive-zone formulation for ITZ with zero-thickness cohesive elements, calibrated with concrete tension and compression experiments. Models omitting ITZ are shown to be in poor agreement with experiments, both qualitatively and quantitatively. Models with finite thickness ITZ are also in poorer agreement with experiments compared to those with zero-thickness, despite higher computational complexity. It is recommended that concrete analyses follow the proposed strategy for meso-structure modelling and constituents’ calibration.