Mesoscale modelling of single-aggregate debonding behaviour in asphalt pavement using the combined finite and discrete element method

Abstract

Aggregate loss from artificial asphalt pavement surfaces widely happens, which causes premature deterioration of roads. However, the debonding mechanism of geographical aggregate is still unrevealed owing to the intricate aggregate morphology and the multiphase structure. To address this challenge, a combined finite and discrete element method (FDEM) was presented to simulate the debonding behaviour in the mesoscale. A mesh-size sensitivity study was first conducted on the asphalt composite to ensure the model self-consistency. Then a realistic particle shape with natural morphology from X-ray micro-computed tomography (XCT) was imported into the numerical model. To explore the role of the aggregate shape on debonding behaviour, systematic simulations were also performed on spherical, ellipsoidal particles for comparison. Results show that the curvature variation could significantly enhance the nesting effect, facilitating stress diffusion. Meanwhile, the further enrichment of irregular curvature variation on real particles results in a more uniform distribution of failure area concerning the aggregate and interface, however, reduces the maximum reaction moment for debonding. The breaking moments for both ellipsoidal and realistic aggregates obey the Generalized extreme value (GEV) distribution. This study helps improve the cognition of aggregate shape effect on debonding behaviour and facilitates the design optimization of the asphalt wearing course.