Investigation of mineral-scale heterogeneity effects on dynamic compression failure behavior of rocks using grain-based discontinuous deformation analysis

Résumé fourni par la source

The increasing scale and complexity of engineering projects have made understanding the dynamic mechanical behavior of rocks crucial for scenarios such as blasting and impact. To investigate the role of mineral-scale heterogeneity in dynamic compression failure, this study develops a grain-based discontinuous deformation analysis (GB-DDA) model using Voronoi tessellation and calibrates it against experimental data for Barre granite. By employing a refined loading protocol, the model incorporates grain-scale heterogeneity into the DDA framework, thus enabling realistic simulation of multi-mineral rocks under dynamic loading while mitigating stress nonuniformity and premature boundary failures in high-strain-rate uniaxial compression tests. By applying varying loading rates to models with different average grain sizes, grain roundness, and mineral contents, we reveal that increasing strain rate intensifies fragmentation, promotes shear and intragranular (TG) cracking, and enhances both dynamic uniaxial compressive strength ( DUCS ) and the dynamic strength increase factor ( DIF ). Furthermore, at constant strain rates, larger grain size, higher grain roundness, and greater biotite content facilitate shear crack initiation, while larger grain size, higher roundness, and greater quartz content promote intragranular cracking and further increase DUCS . Larger grain size and lower roundness elevate DIF , whereas mineral composition has a limited effect on DIF . Two-way analysis of variance (ANOVA) further revealed that, within the tested ranges, mesostructure-strain rate interactions vary by parameter type, with synergistic effects primarily governing shear-dominated failure and strength rather than intragranular cracking.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.