ABSTRACT
This work numerically evaluates the role of material heterogeneity in the variability of fracture processes in Hot Mix Asphalt (HMA) materials. Specifically, the paper focuses on the heterogeneity induced by the spatial air void distribution within HMA specimens, and it uses a fracture test configuration to assess the differences produced by these distributions in the fracture response of the mixtures. The work initiated with an experimental stage that provided fracture, creep compliance and indirect tensile strength (IDT) test results and data on the heterogeneous distribution of air voids (AV) in HMA specimens. The outcome from this phase revealed that the air void content in HMA specimens significantly impacted the viscoelastic, tensile resistance and fracture response of the material. This stage also provided information on the typical internal distribution of AV within cylindrical HMA testing specimens. These experimental results were used in the development of computational Finite Element (FE) models that used the geometrical configuration of the semi-circular bending (SCB) test, and that incorporated cohesive surface contacts and stochastic Random Fields (RF) techniques. The cohesive contacts were used to simulate fracture within the computational specimens, while RF theory was used to indirectly account for the non-uniform spatial distributions of the air void phase and of the mechanical- and fracture-related material properties within the specimens. The numerical results suggest that HMA heterogeneity due to non-uniform air void distributions accounts for an important portion of the variability associated with fracture processes in HMA materials.
Disclosure statement
No potential conflict of interest was reported by the authors.
Acknowledgements
The authors would like to acknowledge Mr. Ken Rutabana for the experimental work conducted at University of Arkansas. The authors are also grateful to Dr. Raj Dongre, who permitted us to use specimens fabricated in his laboratory to conduct most of the experimental portion of this work and to Mr. Oliver Giraldo-Londoño for his support in the numerical implementation of the fracture model. This publication was partially made possible by the call for proposals ‘Research Program 2012’ from the Office of the Vice President for Research at Universidad de Los Andes (Bogota, Colombia). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the University.