TY - JOUR
T1 - Micromechanical modeling of I-FIT asphalt concrete specimens
AU - Hernandez, Jaime
AU - Sawalha, Mohammed
AU - Rivera-Perez, Jose
AU - Ozer, Hasan
AU - Al-Qadi, Imad L.
N1 - Funding Information: The authors would like to acknowledge the financial support of the Center for Highway Pavement Preservation and the US Department of Transportation-Research and Innovative Technology Administration (RITA). Publisher Copyright: © 2018 Elsevier Ltd Copyright: Copyright 2018 Elsevier B.V., All rights reserved.
PY - 2018/9
Y1 - 2018/9
N2 - Analytical and numerical micromechanical models have been previously used to understand fracture behavior of heterogeneous materials like Portland cement concrete or asphalt concrete (AC). In this study, the behavior of asphalt concrete was studied during a semi-circular bending (SCB) fracture test, Illinois Flexibility Index Test (I-FIT), using micromechanical level finite element models. The models were validated in multiple steps using the strain fields calculated with the digital image correlation (DIC) technique as well as the global scale forces measured in the same experiment. The micromechanical model was developed to evaluate the effects of microstructural features such as aggregate gradation, aggregate distribution, and void space on fracture behavior of AC. The model focused on pre-peak behavior and assumed that AC consists of aggregates and mortar. Aggregates were considered linear elastic with material constants reported in the literature, while the mortar was assumed linear viscoelastic. Mortar was defined as the combination of binder, air voids, and material passing 2.36 mm sieve. Mixture theory was utilized to characterize the mortar as viscoelastic using binder's dynamic shear rheometer results, elastic properties of fine aggregate material, and air voids volume from the AC mix design. The validated FEM was used to perform a parametric study aimed at determining the effect of aggregate gradation and binder properties on the applied load, opening strains and stresses, and energy around the crack tip. Nine aggregate gradations and three binders were studied; ten replicates for each aggregate gradation-binder combination were considered. In order to create the replicates, a Python script that fabricates artificial aggregate gradations and randomly distributes aggregates in the I-FIT geometry was created. It was found that mortar properties, rather than air voids, binder content, or fine material, were heavily correlated to energy and applied load of the I-FIT specimen.
AB - Analytical and numerical micromechanical models have been previously used to understand fracture behavior of heterogeneous materials like Portland cement concrete or asphalt concrete (AC). In this study, the behavior of asphalt concrete was studied during a semi-circular bending (SCB) fracture test, Illinois Flexibility Index Test (I-FIT), using micromechanical level finite element models. The models were validated in multiple steps using the strain fields calculated with the digital image correlation (DIC) technique as well as the global scale forces measured in the same experiment. The micromechanical model was developed to evaluate the effects of microstructural features such as aggregate gradation, aggregate distribution, and void space on fracture behavior of AC. The model focused on pre-peak behavior and assumed that AC consists of aggregates and mortar. Aggregates were considered linear elastic with material constants reported in the literature, while the mortar was assumed linear viscoelastic. Mortar was defined as the combination of binder, air voids, and material passing 2.36 mm sieve. Mixture theory was utilized to characterize the mortar as viscoelastic using binder's dynamic shear rheometer results, elastic properties of fine aggregate material, and air voids volume from the AC mix design. The validated FEM was used to perform a parametric study aimed at determining the effect of aggregate gradation and binder properties on the applied load, opening strains and stresses, and energy around the crack tip. Nine aggregate gradations and three binders were studied; ten replicates for each aggregate gradation-binder combination were considered. In order to create the replicates, a Python script that fabricates artificial aggregate gradations and randomly distributes aggregates in the I-FIT geometry was created. It was found that mortar properties, rather than air voids, binder content, or fine material, were heavily correlated to energy and applied load of the I-FIT specimen.
KW - Asphalt concrete
KW - Digital image correlation
KW - Finite elements
KW - Fracture
KW - Illinois flexibility index
KW - Micromechanical modeling
KW - Semi-circular beam
UR - http://www.scopus.com/inward/record.url?scp=85050886095&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85050886095&partnerID=8YFLogxK
U2 - 10.1016/j.engfracmech.2018.07.033
DO - 10.1016/j.engfracmech.2018.07.033
M3 - Article
SN - 0013-7944
VL - 200
SP - 234
EP - 250
JO - Engineering Fracture Mechanics
JF - Engineering Fracture Mechanics
ER -