Asphalt mastic, which consists of an asphalt binder and a mineral filler, provides critical adhesion and viscoelasticity to an asphalt mixture. The rheological response of the asphalt mastic is mainly derived from its asphalt binder. In this study, a simple laboratory test method is proposed to estimate the relationship of asphalt binder and its mastic. Two modified binders (3.5% and 4.0% styrene–butadiene–styrene (SBS) of asphalt binder by mass) were blended with a limestone filler at six different mineral filler contents to produce mastic samples. A modified multiple stress creep-recovery (MSCR) test was conducted on both the asphalt binder and its mastic with the same testing protocols, and the stress conditions and rheological response of asphalt binder in the mastic with linear or nonlinear viscoelasticity were both investigated. The results show that the stress of the asphalt binder in its mastic decreased with increasing filler contents. However, for the linear-viscoelasticity mastic, the decrease rate of the stress began to slow down when the filler content had reached 100% or 120%. For the rheological properties of the asphalt binder in the mastic, the %R of the asphalt binder was improved by adding filler, especially for the nonlinear-viscoelasticity mastic. The asphalt binder of the linear-viscoelasticity asphalt mastic also showed a linear viscoelastic response and a good recovery property. The performance of the asphalt mastic and rheological properties of its asphalt binder were highly related to its filler content.

The mechanical behavior of asphalt mixtures at high stress levels are characterized by non-linear viscoelasticity and damage evolution. A nonlinear damage constitutive model considering the existence of creep hardening and creep damage mechanisms in the entire creep process is proposed in this study by adopting the fractional rheology theory to characterize the three-stage creep process of mixtures. A series of uniaxial compressive creep tests under various stresses were conducted at different temperatures to verify the model. The results indicated that the model predictions were in good agreement with the creep tests. The relationship between the model parameters and applied stresses was established, and the stress range in which the mixture exhibited only creep consolidation was obtained. The damage to the asphalt mixture was initiated in the steady stage; however, it developed in the tertiary stage. A two-parameter Weibull distribution function was used to describe the evolution between the damage values and damage strains at different stress levels and temperatures. The correlation coefficients were greater than 0.99 at different temperatures, indicating that a unified damage evolution model could be established. Thus, the parameters of the unified model were related to material properties and temperature, independent of the stress levels applied to the mixtures.