Experimental investigation and multi-scale Mori–Tanaka modeling of viscoelastic asphalt mastic with imperfect interfaces

Asphalt mastics, comprising asphalt binder and mineral filler, play a critical role in determining the performance of asphalt mixtures. This study examines the viscoelastic behavior of mastics composed of two asphalt binders and three fillers (gabbro, quartz, and hydrated lime) with varying particle sizes. Repeated Creep and Recovery Tests (RCRT) were conducted to evaluate the influence of filler type and concentration on mastic strain response. Then, a modified multi-scale Mori–Tanaka model was successfully employed to predict the viscoelastic properties of the mastics. The model parameters were found to be influenced by the filler type and size, as well as by imperfections at the binder–filler interface. Results show that mastics containing hydrated lime exhibit higher moduli, particularly at high filler content (40% by volume), with significantly lower final strain in RCRT compared to those with gabbro or quartz. This enhanced performance is attributed to the high surface area of hydrated lime and its chemical interactions with the asphalt binder, which promote the formation of a network structure within the mastic. While the proposed model effectively captures the physical interactions in asphalt mastics, it requires further refinement to account for the additional influence of chemical reactions between fillers, such as hydrated lime, and the asphalt binder. • Hydrated lime yields the highest stiffness through strong physio-chemical interactions with the binder. • Filler effects are minimal at low concentrations (10%) but enhance stiffness at higher levels (30%). • The model captures gabbro/quartz behavior well but poorly captures hydrated lime’s chemical effects. • Model parameters ‘a’ and λ reflect the effects of filler concentration and interfacial bonding.