Characterization of Asphalt Mixtures Using Controlled-Strain Repeated Direct Tension Test

Abstract

A controlled-strain repeated direction tension (RDT) test is useful for characterizing asphalt mixtures and assessing fatigue cracking accompanied by plastic deformation. The strain curve in a loading cycle of the controlled-strain RDT test is a standard haversine wave with only a tensile portion, whereas the stress curve has a tensile portion and a compressive portion (quasi-compression) because of the viscoelastoplastic nature of the asphalt mixture. Usually the material properties in the tensile stress portion (the tensile properties) are used, while those in the compressive stress portion (the quasi-compressive properties) are neglected. This leads to nonnegligible errors in estimating the dissipated strain energy (DSE) and characterizing the damage (such as cracking and permanent deformation) generated in the material. The objective of this paper is to obtain complete material properties and accurate amounts of dissipation and storage of energy of this test method as the foundation of damage characterization. This paper provides methods to simulate the stress and strain for the tensile and compressive stress portions separately so as to determine the tensile and quasi-compressive properties, respectively. The difference between the tensile and quasi-compressive properties is investigated using statistical analysis. It is found that the tensile properties of an asphalt mixture are different from the quasi-compressive properties due to the crack opening and closure within each loading cycle. The DSE calculated using both the tensile and quasi-compressive properties is larger compared to the traditional method using only the tensile properties, indicating an underestimation of the damage by the traditional method. In addition, the recoverable strain energy (RSE) is investigated to examine the ability of an asphalt mixture to store and recover energy. A new perspective in terms of the integration method is provided to determine the actual amount of the RSE in a loading cycle.